28-02-2013, 09:33 AM
Static and Dynamic Characteristics of Instrumentation
Static and Dynamic.pdf (Size: 394.74 KB / Downloads: 64)
INTRODUCTION
Before we can begin to develop an understanding of the static and time changing characteristics of
measurements, it is necessary to build a framework for understanding the process involved, setting down
the main words used to describe concepts as we progress.
Measurement
is the process by which relevant information about a system of interest is interpreted
using the human thinking ability to define what is believed to be the new knowledge gained. This
information may be obtained for purposes of controlling the behavior of the system (as in engineering
applications) or for learning more about it (as in scientific investigations).
The basic entity needed to develop the knowledge is called
data
, and it is obtained with physical
assemblies known as sensors that are used to observe or sense system variables. The terms
information
and
knowledge
tend to be used interchangeably to describe the entity resulting after data from one or
more sensors have been processed to give more meaningful understanding. The individual variables being
sensed are called
measurands
The most obvious way to make observations is to use the human senses of seeing, feeling, and hearing.
This is often quite adequate or may be the only means possible. In many cases, however, sensors are used
that have been devised by man to enhance or replace our natural sensors. The number and variety of
sensors is very large indeed. Examples of man-made sensors are those used to measure temperature,
pressure, or length. The process of sensing is often called
transduction
, being made with transducers.
These man-made sensor assemblies, when coupled with the means to process the data into knowledge,
are generally known as (measuring) instrumentation.
Static Characteristics of Instrument Systems
Output/Input Relationship
Instrument systems are usually built up from a serial linkage of distinguishable building blocks. The
actual physical assembly may not appear to be so but it can be broken down into a representative diagram
of connected blocks. Figure 3.4 shows the block diagram representation of a humidity sensor. The sensor
is activated by an input physical parameter and provides an output signal to the next block that processes
the signal into a more appropriate state.
A key generic entity is, therefore, the relationship between the input and output of the block. As was
pointed out earlier, all signals have a time characteristic, so we must consider the behavior of a block in
terms of both the static and dynamic states.
Drift
It is now necessary to consider a major problem of instrument performance called
instrument drift
. This
is caused by variations taking place in the parts of the instrumentation over time. Prime sources occur
as chemical structural changes and changing mechanical stresses. Drift is a complex phenomenon for
which the observed effects are that the sensitivity and offset values vary. It also can alter the accuracy of
the instrument differently at the various amplitudes of the signal present.
Detailed description of drift is not at all easy but it is possible to work satisfactorily with simplified
values that give the average of a set of observations, this usually being quoted in a conservative manner.
The first graph (
a
) in Figure 3.6 shows typical steady drift of a measuring spring component of a weighing
balance. Figure 3.6(
b
) shows how an electronic amplifier might settle down after being turned on.
Drift is also caused by variations in environmental parameters such as temperature, pressure, and
humidity that operate on the components. These are known as
influence parameters
. An example is the
change of the resistance of an electrical resistor, this resistor forming the critical part of an electronic
amplifier that sets its gain as its operating temperature changes.
Unfortunately, the observed effects of influence parameter induced drift often are the same as for time
varying drift. Appropriate testing of blocks such as electronic amplifiers does allow the two to be separated
to some extent. For example, altering only the temperature of the amplifier over a short period will
quickly show its temperature dependence.
Hysteresis and Backlash
Careful observation of the output/input relationship of a block will sometimes reveal different results as
the signals vary in direction of the movement. Mechanical systems will often show a small difference in
length as the direction of the applied force is reversed. The same effect arises as a magnetic field is reversed
in a magnetic material. This characteristic is called
hysteresis
. Figure 3.7 is a generalized plot of the
output/input relationship showing that a closed loop occurs. The effect usually gets smaller as the
amplitude of successive excursions is reduced, this being one way to tolerate the effect. It is present in
most materials. Special materials have been developed that exhibit low hysteresis for their application —
transformer iron laminations and clock spring wire being examples.
Where this is caused by a mechanism that gives a sharp change, such as caused by the looseness of a
joint in a mechanical joint, it is easy to detect and is known as
backlash
Saturation
So far, the discussion has been limited to signal levels that lie within acceptable ranges of amplitude. Real
system blocks will sometimes have input signal levels that are larger than allowed. Here, the dominant
errors that arise —
saturation
and
crossover distortion
— are investigated.
As mentioned above, the information bearing property of the signal can be carried as the instantaneous
value of the signal or be carried as some characteristic of a rapidly varying ac signal. If the signal form
is not amplified faithfully, the output will not have the same linearity and characteristics.
The gain of a block will usually fall off with increasing size of signal amplitude. A varying amplitude
input signal, such as the steadily rising linear signal shown in Figure 3.8, will be amplified differently
according to the gain/amplitude curve of the block. In uncompensated electronic amplifiers, the larger
amplitudes are usually less amplified than at the median points.