MODULE 3

PRINCIPLES OF MEASUREMENT

PERFORMANCE CHARACTERISTICS

Knowledge of the performance characteristics of an instrument is essential for selecting the most
suitable instrument for specific measuring jobs. It consists of two basic characteristics-static and
dynamic.

STATIC CHARACTERISTICS

The static characteristics of an instrument are, in general, considered for instruments which are
used to measure an unvarying process condition. All the static performance characteristics are
obtained by one form or another of a process called calibration. There are a number of related
definitions (or characteristics), which are described below, such as accuracy, precision,
repeatability, resolution, errors, sensitivity, etc.

1. Instrument
A device or mechanism used to determine the present value of the quantity under
measurement.
2. Measurement
The process of determining the amount, degree, or capacity by comparison (direct or
indirect) with the accepted standards of the system units being used.
3. Accuracy
The degree of exactness (closeness) of a measurement compared to the expected (desired)
value.

4. Resolution

The smallest change in a measured variable to which an instrument will respond.

5. Precision

A measure of the consistency or repeatability of measurements, i.e. successive reading do

The design value, i.e. the most probable value that calculations indicate one should expect
to measure.

7. Error

The deviation of the true value from the desired value.

8. Sensitivity

The ratio of the change in output (response) of the instrument to a change of input or
measured variable.

ERROR IN MEASUREMENT

Measurement is the process of comparing an unknown quantity with an accepted standard quantity.
It involves connecting a measuring instrument into the system under consideration and observing
the resulting response on the instrument. The measurement thus obtained is a quantitative measure
of the so-called "true value" (since it is very difficult to define the true value, the term "expected
value" is used). Any measurement is affected by many variables; therefore, the results rarely reflect
the expected value. For example, connecting a measuring instrument into the circuit under
consideration always disturbs (changes) the circuit, causing the measurement to differ from the
expected value. Some factors that affect the measurements are related to the measuring instruments
themselves. Other factors are related to the person using the instrument. The degree to which a
measurement nears the expected value is expressed in terms of the error of measurement. Error
may be expressed either as absolute or as percentage of error.

Absolute error may be defined as the difference between the expected value of the variable and
the measured value of the variable, or
GROSS ERRORS

These errors are mainly due to human mistakes in reading or in using instruments or errors in
recording observations. Errors may also occur due to incorrect adjustment of instruments and
computational mistakes. These errors cannot be treated mathematically. The complete elimination
of gross errors is not possible, but one can minimize them. Some errors are easily detected while
others may be elusive. One of the basic gross errors that occurs frequently is the improper use of
an instrument. The error can be minimized by taking proper care in reading and recording the
measurement parameter. In general, indicating instruments change ambient conditions to some
extent when connected into a complete circuit.
Systematic Errors

These errors occur due to shortcomings of the instrument, such as defective or worn parts, or
ageing or effects of the environment on the instrument.

These errors are sometimes referred to as bias, and they influence all measurements of a quantity
alike. A constant uniform deviation of the operation of an instrument is known as a systematic
error. There are basically three types of systematic errors-(i) Instrumental, (ii) Environmental, and
(iii) Observational.

(I) INSTRUMENTAL ERRORS Instrumental errors are inherent in measuring instruments,

Instrumental errors can be avoided by (a) selecting a suitable instrument for the particular
measurement applications. (Refer Examples 1.3 (a) and (bj). (b) applying correction factors after
determining the amount of instrumental error. (c) calibrating the instrument against a standard

(II) ENVIRONMENTAL ERROR

Environmental errors are due to conditions external to the measuring device, including conditions
in the area surrounding the instrument, such as the effects of change in temperature, humidity,
barometric pressure or of magnetic or electrostatic fields. These errors can also be avoided by (i)
air conditioning, (ii) hermetically sealing certain components in the instruments, and (iii) using
magnetic shields.

(III) OBSERVATIONAL ERRORS

Observational errors are errors introduced by the observer. The most common error is the parallax
error introduced in reading a meter scale, and the error of estimation when obtaining a reading
from a meter scale. These errors are caused by the habits of individual observers. For example, an
observer may always introduce an error by consistently holding his head too far to the left while
reading a needle and scale reading. In general, systematic errors can also be subdivided into static
and dynamic errors. Static errors are caused by limitations of the measuring device or the physical
laws governing its behavior. Dynamic errors are caused by the instrument not responding fast
enough to follow the changes in a measured variable.

Example 1.3 (a) A voltmeter having a sensitivity of 1 kWV is connected across an unknown
resistance in series with a milliammeter reading 80 Von 150 Vscale. When the milliammeter reads
10 mA, calculate the (i) Apparent resistance of the unknown resistance, (ii) Actual resistance of
the unknown resistance, and (iii) Error due to the loading effect of the voltmeter
n Example 1.3 (a), a well calibrated voltmeter may give a misleading resistance when connected
across two points in a high resistance circuit. The same voltmeter, when connected in a low
resistance circuit (Example 1.3 (b) may give a more dependable reading. This show that voltmeters
have a loading effect in the circuit during measurement,

Random Errors

These are errors that remain after gross and systematic errors have been substantially reduced or
at least accounted for. Random errors are generally an accumulation of a large number of small
effects and may be of real concern only in measurements requiring a high degree of accuracy. Such
errors can be analyzed statistically. These errors are due to unknown causes, not determinable in
the ordinary process of making measurements. Such errors are normally small and follow the laws
of probability. Random errors can thus be treated mathematically. For example, suppose a voltage
is being monitored by a voltmeter which is read at 15 minutes intervals. Although the instrument
operates under ideal environmental conditions and is accurately calibrated before measurement, it
still gives readings that vary slightly over the period of observation. This variation cannot be
corrected by any method of calibration or any other known method of control.

MULTI RANGE AMMETERS

The current range of the dc ammeter may be further extended by a number of shunts, selected by
a range switch, such a meter is called a multirange ammeter, shown in Fig. The circuit has four
shunts RI, R2, R3 and R4, which can be placed in parallel with the movement to give four different
current ranges. Switch S is a multiposition switch, (having low contact resistance and high current
carrying capacity, since its contacts are in series with low resistance shunts). Make before break
type switch is used for range changing. This switch protects the meter movement from being
damaged without a shunt during range changing. If we use an ordinary switch for range changing,
the meter does not have any shunt in parallel while the range is being changed, and hence full
current passes through the meter movement, damaging the movement. Hence a make before break
type switch is used. The switch is so designed that when the switch position is changed, it makes
contact with the next terminal (range) before breaking contact with the previous terminal.
Therefore the meter movement is never left unprotected. Multirange ammeters are used for ranges
up to 50A. When using a multirange ammeter, first use the highest current range, then decrease the
range until good upscale reading is obtained. The resistance used for the various ranges are of very
high precision values, hence the cost of the meter increases
Multirange Voltmeter:

As in the case of an ammeter, to obtain a multirange ammeter, a number of shunts are connected
across the movement with a multi-position switch. Similarly, a dc voltmeter can be converted into
a multirange voltmeter by connecting a number of resistors (multipliers) along with a range switch
to provide a greater number of workable ranges.
Figure 4.2 shows a multirange voltmeter using a three-position switch and three multipliers R1, R2,
and R3 for voltage values V1, V2, and V3. Figure 4.2 can be further modified to Fig. 4.3, which is
a more practical arrangement of the multiplier resistors of a multirange voltmeter.

In this arrangement, the multipliers are connected in a series string, and the range selector selects
the appropriate amount of resistance required in series with the movement.

This arrangement is advantageous compared to the previous one, because all multiplier resistances
except the first have the standard resistance value and are also easily available in precision
tolerances:

The first resistor or low range multiplier, R4, is the only special resistor which has to be specially
manufactured to meet the circuit requirements.

Digital voltmeters (DVMs) are measuring instruments that convert analog voltage signals into a
digital or numeric readout. This digital readout can be displayed on the front panel and also used
as an electrical digital output signal. Any DVM is capable of measuring analog de voltages.
However, with appropriate signal conditioners preceding the input of the DVM, quantities such as
ac voltages, ohms, dc and ac current, temperature, and pressure can be measured. The common
element in all these signal conditioners is the de voltage, which is proportional to the level of the
unknown quantity being measured. This de output is then measured by the DVM. DVMs have
various features such as speed, automation operation and programability.

There are several varieties of DVM which differ in the following ways:

1. Number of digits

2. Number of measurements

3. Accuracy

4. Speed of reading

5. Digital output of several types.

The DVM displays ac and de voltages as discrete numbers, rather than as a pointer on a continuous
scale as in an analog voltmeter. A numerical readout is advantageous because it reduces human
error, eliminates parallax error, increases reading speed and often provides output in digital form
suitable for further processing and recording. With the development of IC modules, the size, power
requirements and cost of DVMs have been reduced, so that DVMS compete with analog voltmeters
in portability and size. Their outstanding qualities are their operating and performance
characteristics, as detailed below.

1. Input range from + 1.000 V to + 1000 V with automatic range selection and overload indication
2. Absolute accuracy as high as ±0.005% of the reading

3. Resolution 1 part in million (1 Jl,Vreading can be read or measured on 1 V range)

4. Input resistance typically 10 MQ, input capacitance 40 pF

5. Calibration internally from stabilised reference sources, independent of measuring circuit

6. Output in BCD form, for print output and further digital processing.

Optional features may include additional circuitry to measure current, ohms and voltage ratio.
DUAL SLOPE INTEGRATING TYPE DVM (VOLTAGE TO TIME
CONVERSION)

In ramp techniques, superimposed noise can cause large errors. In the dual ramp technique, noise
is averaged out by the positive and negative ramps using the process of integration.

Principle of Dual Slope Type DVM As illustrated in Fig. 5.3, the input voltage 'e/ is integrated,
with the slope of the integrator output proportional to the test input voltage. After a fixed time,
equal to t., the input voltage is disconnected and the integrator input is connected to a negative
voltage -er. The integrator output will have a negative slope which is constant and proportional to
the magnitude of the input voltage. The block diagram is given in Fig. 5.4.
At the start a pulse resets the counter and the FIF output to logic level '0'. S, is closed and S, is
open. The capacitor begins to charge. As soon as the integrator output exceeds zero, the comparator
output voltage changes state, which opens the gate so that the oscillator clock pulses are fed to the
counter. (When the ramp voltage starts, the comparator goes to state 1, the gate opens and clock
pulse drives the counter.) When the counter reaches maximum count, i.e. the counter is made to
run for a time 't)' in this case 9999, on the next clock pulse all digits go to 0000 and the counter
activates the FIF to logic level '1'. This activates the switch drive, ei is disconnected and -er is
connected to the integrator. The integrator output will have a negative slope which is constant, i.e.
integrator output now decreases linearly to 0 volts. Comparator output state changes again and
locks the gate. The discharge time t2 is now proportional to the input voltage. The counter indicates
the count during time t2• When the negative slope of the integrator reaches zero, the comparator
switches to state 0 and the gate closes, i.e. the capacitor C is now discharged with a constant slope.
As soon as the comparator input (zero detector) finds that eo is zero, the counter is stopped. The
pulses counted by the counter thus have a direct relation with the input voltage. During charging
From Eq. 5.3 it is evident that the accuracy of the measured voltage is independent of the integrator
time constant. The times tI and t2 are measured by the count of the clock given by the numbers nl
and n2 respectively. The clock oscillator period equals T and if n, and e, are constants, then Eq.
5.4 indicates that the accuracy of the method is also independent of the oscillator frequency. The
dual slope technique has excellent noise rejection because noise and superimposed ac are averaged
out in the process of integration. The speed and accuracy are readily varied according to specific
requirements; also an accuracy of ± 0.05% in 100 ms is available.

Direct Compensation:

The input signal is compared with an internally generated voltage which is increased in steps
starting from zero. The number of steps needed to reach the full compensation is counted. A simple
compensation type is the staircase ramp

The Staircase Ramp The basic principle is that the input signal ~ is compared with an internal
staircase voltage, Vc> generated by a series circuit consisting of a pulse generator (clock), a
counter counting the pulses and a digital to analog converter, converting the counter output into a
de signal. As soon as Vc is equal to ~, the input comparator closes a gate between the clock and
the counter, the counter stops and its output is shown on the display. The basic block diagram is
shown in Fig.
Operation of the Circuit The clock generates pulses continuously. At the start of a measurement,
the counter is reset to 0 at time f 1 so that the output of the digital to analog converter (DAC) is
also O.If ~ is not equal to zero, the input comparator applies an output voltage that opens the gate
so that clock pulses are passed on to the counter through the gate. The counter starts counting and
the DAC starts to produce an output voltage increasing by one small step at each count of the
counter. The result is a staircase voltage applied to the second input of the comparator, as shown
in Fig.

This process continues until the staircase voltage is equal to or slightly greater than the input
voltage ~. At that instant f2' the output voltage of the input comparator changes state or polarity,
so that the gate closes and the counter is stopped.

The display unit shows the result of the count. As each count corresponds to a constant dc step in
the DAC output voltage, the number of counts is directly proportional to Vc and hence to Vi. By
appropriate choice of reference voltage, the step height of the staircase voltage can be determined.
For example, each count can represent l mV and direct reading of the input voltage in volts can be
realised by placing a decimal point in front of the 10 decade.

The advantages of a staircase type DVM are as follows: 1. Input impedance of the DAC is high
when the compensation is reached. 2. The accuracy depends only on the stability and accuracy of
the voltage and DAC. The clock has no effect on the accuracy.

The disadvantages are the following: 1. The system measures the instantaneous value of the input
signal at the moment compensation is reached. This means the reading is rather unstable, i.e. the
input signal is not a pure de voltage. 2. Until the full compensation is reached the input impedance
is low, which can influence the accuracy.

SUCCESSIVE APPROXIMATION

The successive approximation principle can be easily understood using a simple example; the
determination of the weight of an object. By using a balance and placing the object on one side
and an approximate weight on the other side, the weight of the object is determined. If the weight
placed is more than the unknown weight, the weight is removed and another weight of smaller
value is placed and again the measurement is performed. Now if it is found that the weight placed
is less than that of the object, another weight of smaller value is added to the weight already
present, and the measurement is performed. If it is found to be greater than the unknown weight
the added weight is removed and another weight of smaller value is added. In this manner by
adding and removing the appropriate weight, the weight of the unknown object is determined. The
successive approximation DVM works on the same principle. Its basic block diagram is shown in
Fig. 5.10. When the start pulse signal activates the control circuit, the successive approximation
register (SAR) is cleared. The output of the SAR is 00000000. VOU! of the DIA converter is o.
Now, if Vin > VOU! the comparator output is positive. During the first clock pulse, the control
circuit sets the D7 to 1, and Voutjumps to the half reference voltage. The SAR output is 10000000.
If VOll!is greater than Vin, the comparator output is negative and the control circuit resets D7.
However, if Vin is greater than VOU!' the comparator output is positive and the control circuits
keeps D7 set. Similarly the rest of the bits beginning from D7 to Do are set and tested. Therefore,
the measurement is completed in 8 clock pulses.
At the beginning of the measurement cycle, a start pulse is applied to the start-stop multivibrator.
This sets a 1 in the MSB of the control register and a o in all bits (assuming an 8-bit control) its
reading would be 10000000. This initial setting of the register causes the output of the DIA
converter to be half the reference voltage, i.e. 1/2 V.This converter output is compared to the
unknown input by the comparator. If the input voltage is greater than the converter reference
voltage, the comparator output produces an output that causes the control register to retain the 1
setting in its MSB and the converter continues to supply its reference output voltage of 1/2 Vref.
The ring counter then advances one count, shifting a 1 in the second MSB of the control register
and its reading becomes 11000000. This causes the DIA converter to increase its reference output
by 1 increment to 1/4 V, i.e. 1/2 V + 1/4 V, and again it is compared with the unknown input. If in
this case the total reference voltage exceeds the unknown voltage, the comparator produces an
output that causes the control register to reset its second MSB to O.The converter output then
returns to its previous value of 1/2 V and awaits another input from the SAR. When the ring counter
advances by 1, the third MSB is set to 1 and the converter output rises by the next increment of
1/2 V + 1/8 V. The measurement cycle thus proceeds through a series of successive
approximations. Finally, when the ring counter reaches its final count, the measurement cycle stops
and the digital output of the control register represents the final approximation of the unknown
input voltage.

Various output levels for each bit (a-Bit shows the voltage level very nearly equal to I V)

Therefore, Vin nearly equals VOUI' i.e. Vin = 1 Vand Vout = 0.99785. The main advantage of this
method is speed. At best it takes n clock pulses to produce an n bit result. Even if the set, test, set
or reset operation takes more than 1 clock pulse, the SAR method is still considerably faster than
the counter method. However the control circuit is more complex in design and cost is enhanced.
This digital voltmeter is capable of 1000 readings per second. With input voltages greater than de,
the input level changes during digitisation and decisions made during conversion are not
consistent. To avoid this error, a sample and hold circuit is used and placed in the input directly
following the input attenuator and amplifier. Switch Drive In its simplest form, the sample and
hold (S/H) circuit can be represented by a switch and a capacitor, as shown in Fig. 5.12.

Simple sample hold circuit

In the Sample mode, the switch is closed and the capacitor charges to the instantaneous value of
the input voltage. In the Hold mode, the switch is opened and the capacitor holds the voltage that
it had at the instant the switch was opened. If the switch drive is synchronized with the ring counter
pulse, the actual measurement and conversion takes place when the SIR circuit is in the Hold mode.
The output waveform of a sample and hold circuit is shown in Fig. 5.13.

Output wavwform of a sample and hold circuit

An actual sample and hold circuit is shown in Fig. 5.14. The sample pulse operates switches 1 and
3. The hold pulse operates switches 2 and 4. The samplehold pulses are complementary. In the
sample mode the hold capacitor is charged up by the Opamp. Inthe hold mode, the capacitor is
switched into the feedback loop, while input resistors R I and Rfare switched to ground. Opamps
are used to increase the available driving current into the capacitor or to isolate the capacitor from
an external load on the output.
 Practical sample and hold circuit

The SIH circuit is basically an Op amp that charges the capacitor during the Sample mode and
retains the charge during the Hold mode.