Page 48 - Understanding Automotive Electronics

2735 | CH 2 Page 35 Tuesday, March 10, 1998 10:55 AM

THE SYSTEMS APPROACH TO CONTROL AND INSTRUMENTATION 2

In this system, the input is the mechanical vibration of the phonograph
needle as it tracks along the groove in the record. The sensor is the phonograph
cartridge that converts these mechanical vibrations to an analog electrical
signal. This electrical signal, which is too weak to drive the loudspeakers (the
actuators in the present example) at an acceptable audio level, is amplified in
the stereo amplifier. The amplifier increases the power level to a point at which
it can drive the loudspeakers. In mathematical terms, if the power level input to
the amplifier is P , then the power output to the speakers (denoted P ) is an
i
o
amplified version of the input:
P = GP i
o
where G is the power gain of the amplifier. That is, the input power is
continuously amplified by the amplifier by a factor of G.

CHARACTERISTICS OF A DIGITAL ELECTRONIC SYSTEM
In contrast to an analog electronic system that operates in continuous
time, a digital system operates in discrete instants of time. This process of
representing a continuous-time quantity at specific discrete times is called
sampling and is illustrated in Figure 2.4.
Figure 2.4a illustrates a continuously varying quantity that is denoted x
(which might, for example, be intake manifold pressure). This continuous-time
quantity is sampled electronically at times that are multiples of a basic sample
period. Figure 2.4a depicts the sample points of the continuous pressure as asterisks.
Each sample is the value of the continuous variable at a specific (discrete) time. A
sequence of samples is presented to the signal processor at the corresponding sample
times. The sequence of samples is shown in Figure 2.4b. In a digital electronic
system, the signal processing is performed by some form of digital computer. This
computer requires time to perform its computations. The time between samples
provides an interval in which the necessary computations are performed.
The time between any successive samples is normally a constant known as
sample time. Sample time is a critically important parameter for any digital
system and is chosen with great care by the system designer. It must be
sufficiently long to enable the computer to perform its computations on any
given sample before the next sample is taken, or the computer cannot keep up
with the data stream in real time. On the other hand, if the sample time is too
long, then the input might change too much for the sampled data to adequately
represent the continuous quantity being sampled. The time required for
computation on each sample is influenced in part by the processor speed and by
the efficiency of the program being used to perform the computations. This
aspect of performance is discussed in greater detail in Chapter 4.
The sampled data illustrated in Figure 2.4b are in a sampled analog
format. This format is not compatible with a digital system. One more step,
called quantization, is required to convert the sampled analog data into data
that can be read by the computer. In a digital electronic system, each sample is
represented numerically by its magnitude. For example, a sequence of samples

UNDERSTANDING AUTOMOTIVE ELECTRONICS 35