ERROR / UNCERTAINTY

IN MEASUREMENT

Dr. Rohit Singh Lather

INTRODUCTION ( Measurement = Best Value ±Uncertainty) units

• The process of evaluating the uncertainty associated with a measurement result is often called uncertainty
analysis or error analysis.
• Through measurement, we try to obtain the value of an unknown parameter.
• Measured value cannot be the actual or true value.
- measured value is very close to the true value, we call it to be a very accurate measuring system.
• Measured data must be evaluated for uncertainty (how accurate is the data?)
• Error analysis is an integral part of measurement.
• Clear idea what are the sources of error, how they can be reduced by properly designing the measurement
methodology and by repetitive measurements.
• Besides, for maintaining the accuracy the readings of the measuring instrument are frequently to be compared and
adjusted with the reading of another standard instrument. (Calibration)

2
 ERRORS IN MEASURMENT

• Measurement is the process of comparing an unknown quantity with an accepted standard quantity.
• The measurement thus obtained is a quantitative measure of the so- called “true value” or “expected
value”.
• Some factors that affect measurements are related to the measuring instruments themselves.
• Others factors are related to the person using the instrument.
• The degree to which a measurement conforms to the expected value is expressed in terms of error of
the measurement.
• Error in measurement (e) is defined as the algebraic difference between the true value of the
variable (or expected value) and the measured value.
• Error in measurement in others words is the uncertainty of measured values.

20/03/22 3
 NO MEASURMENT IS ACCURATE!
• Parallax error
• incorrectly sighting the measurement
• Calibration error
• if the scale is not accurate
• Zero error
• if the device doesn’t have a zero or isn’t correctly set to zero
• Damage
• if the device is damaged or faulty
• Limit of reading of the measurement device
• the measurement can only be as accurate as the smallest unit of measurement of the device

20/03/22 4
• Precision indicates the quality of the measurement, without any guarantee that the
measurement is “correct.”
• Precision is often reported quantitatively by using relative or fractional uncertainty
𝑈𝑛𝑐𝑒𝑟𝑡𝑎𝑖𝑛𝑡𝑦
𝑅𝑒𝑙𝑎𝑡𝑖𝑣𝑒 𝑈𝑛𝑐𝑒𝑟𝑡𝑎𝑛𝑖𝑡𝑦 =
𝑀𝑒𝑎𝑠𝑢𝑟𝑒𝑑 𝑄𝑢𝑎𝑛𝑡𝑖𝑡𝑦
0.5 𝑔
𝐸𝑥: 𝑀𝑎𝑠𝑠 = 75.5 ± 0.5 𝑔 ℎ𝑎𝑠 𝑎 𝑓𝑟𝑎𝑐𝑡𝑖𝑜𝑛𝑎𝑙 𝑢𝑛𝑐𝑒𝑟𝑡𝑎𝑛𝑖𝑡𝑦 𝑜𝑓: = 0.006? = 0.7%
75.5𝑔

• Accuracy, on the other hand, assumes that there is an ideal value, and tells how far your answer is
from that ideal, “right” answer
• Accuracy is often reported quantitatively by using relative error:

𝑀𝑒𝑎𝑠𝑢𝑟𝑒𝑑 𝑉𝑎𝑙𝑢𝑒 − 𝑒𝑥𝑝𝑒𝑐𝑡𝑒𝑑 𝑣𝑎𝑙𝑢𝑒

𝑅𝑒𝑙𝑎𝑡𝑖𝑣𝑒 𝐸𝑟𝑟𝑜𝑟 =
𝐸𝑥𝑝𝑒𝑐𝑡𝑒𝑑 𝑉𝑎𝑙𝑢𝑒
75.5 − 80
𝐼𝑓 𝑡ℎ𝑒 𝑒𝑥𝑝𝑒𝑐𝑡𝑒𝑑 𝑣𝑎𝑙𝑢𝑒 𝑓𝑜𝑟 𝑚 𝑖𝑠 80𝑔 𝑡ℎ𝑒𝑛 𝑡ℎ𝑒 𝑟𝑒𝑙𝑎𝑡𝑖𝑣𝑒 𝑒𝑟𝑟𝑜𝑟 𝑖𝑠: = − 0.056 = −5.6%
80
5
 𝑇ℎ𝑒 𝑡𝑒𝑟𝑚 𝑒𝑟𝑟𝑜𝑟 𝑖𝑛 𝑎 𝑚𝑒𝑎𝑠𝑢𝑟𝑚𝑒𝑛𝑡 𝑖𝑠 𝑑𝑒𝑓𝑖𝑛𝑒𝑑 𝑎𝑠:

𝑬𝒓𝒓𝒐𝒓 = 𝐼𝑛𝑠𝑡𝑟𝑢𝑚𝑒𝑛𝑡 𝑟𝑒𝑎𝑑𝑖𝑛𝑔 − 𝑇𝑟𝑢𝑒 𝑟𝑒𝑎𝑑𝑖𝑛𝑔

𝑀𝑒𝑎𝑠𝑢𝑟𝑒𝑑 𝑣𝑎𝑙𝑢𝑒 − 𝑇𝑟𝑢𝑒 𝑣𝑎𝑙𝑢𝑒

% 𝑬𝒓𝒓𝒐𝒓 = × 100%
𝑇𝑟𝑢𝑒 𝑉𝑎𝑙𝑢𝑒

𝐴𝑏𝑠𝑜𝑙𝑢𝑡𝑒 𝑒𝑟𝑟𝑜𝑟
𝑃𝑒𝑟𝑐𝑒𝑛𝑡 𝑜𝑓 𝑒𝑟𝑟𝑜𝑟 = × 100%
𝑇𝑟𝑢𝑒 𝑉𝑎𝑙𝑢𝑒
𝐴𝑏𝑠𝑜𝑙𝑢𝑡𝑒 𝑒𝑟𝑟𝑜𝑟
𝑅𝑒𝑎𝑙𝑖𝑣𝑒 𝐴𝑐𝑐𝑢𝑟𝑎𝑐𝑦 = 1 −
𝑇𝑟𝑢𝑒 𝑉𝑎𝑙𝑢𝑒 Error’ is defined as the difference between the
measured and the true value (as per standard).
g = 9.8 m/sec2
The measured values are enclosed in the error
Measured: g = 9.7 m/sec2 bands, representing the precision of
measurement

Relative error =
6
• Absolute error in a quantity simply means the direct amount by which the given quantity may be
greater or smaller.
• The absolute error is the same units as the measured value and must be expressed in consistent units: (1.500
± .001) m or 1.500 m ± 1 mm are equivalent acceptable forms.
• Both establish a range 1.499 to 1.501 m which probably includes the correct value.
• Note that 1.500 m ± .001 is not a proper expression.
• Relative error is the error expressed as a fraction of the quantity itself, usually in terms of percent (=
parts per hundred), sometimes parts per thousand, and in more accurate work parts per million,
abbreviated ppm.
• The measurement 3.00 m ± 1% is identical to (3.00 ± 0.03) m.
• Relative error itself is dimensionless.

20/03/22 7
 Systematic Random or Accidental Gross Errors
• Instrument • Inconsistencies associated • Personal/human
with accurate measurement of
small quantities
• Environmental • Presence of certain system • Errors due to faulty
defects components
• Loading • Effect of randomly varying • Improper application of the
parameters instrument

• Gross errors: arise due to human mistakes, such as, reading of the instrument value before it reaches steady
state, mistake of recording the measured data in calculating a derived measured, etc. Parallax error in reading on
an analog scale is also is also a source of gross error. Careful reading and recording of the data can reduce the
gross errors to a great extent.
• Systematic errors: are those that affect all the readings in a particular fashion. Zero error, and bias of an
instrument are examples of systematic errors. systematic errors can be corrected by calibration.
• Random error: the cause of which is not clearly known, and they affect the readings in a random way. Random
errors can never be corrected, the can only be reduced by averaging, or error limits can be estimated.
8
 SYSTEMATIC ERRORS

• Systematic errors may arise due to different reasons.

• Shortcomings of the instrument or the sensor
• An instrument may have a zero error, or its output may be varying in a nonlinear fashion with the input, thus
deviating from the ideal linear input/output relationship.
• It may arise due to the loading effect, improper selection of the sensor or the filter cut off frequency.
• Systematic errors can be due to environmental effect also. The sensor characteristics may change with
temperature or other environmental conditions.
• Systematic error can also be due to improper design of the measuring scheme.
Ex : The amplifier inside the instrument may have input offset voltage and current which will contribute to zero
error. Different nonlinearities in the amplifier circuit will also cause error due to nonlinearity.

• The major feature of systematic errors is that the sources of errors are recognizable and can be reduced
to a great extent by carefully designing the measuring system and selecting its components.
9
• By placing the instrument in a controlled environment may also help in reduction of systematic errors.
• They can be further reduced by proper and regular calibration of the instrument.
• Instruments errors: errors due to changes in the proprieties of the components used in instrument such
as friction in the bearing of the meter movement, incorrect spring tension, improper calibration, or faulty
instruments. Instruments errors can be reduced by proper maintenance, use, and handling of
instruments.
• Environmental errors: such errors are due to the environmental conditions in which instruments may
be used. Subjecting instruments to harsh environment such as high temperatures, pressure, or humidity;
strong electrostatic and/or electromagnetic fields, may have detrimental effects, thereby causing errors.
• Observational errors: errors introduced by the observer. Probably the most common observational
errors are the parallax error introduced in reading a meter scale and the error of estimation when
obtaining a reading from a meter scale. 10
 RANDOM ERROR

• Random errors: These are errors that remain after the gross and systematic errors have been substantially
reduced, or at least accounted for.
• Random errors are generally the accumulation of a large number of small effects and may be of real concern only
in measurements requiring a high degree of accuracy.
• Can only be analysed statistically.
• Causes of random errors are not exactly known, so they cannot be eliminated.
• They can only be reduced, and the error ranges can be estimated by using some statistical operations.
• If we measure the same input variable a number of times, keeping all other factors affecting the measurement
same, the same measured value would not be repeated, the consecutive reading will rather differ in a random way.
• But fortunately, the deviations of the readings normally follow a particular distribution (mostly normal distribution),
and we may be able to reduce the error by taking a number of readings and averaging them out.

11
 LIMITING ERROR

• Limiting error is an important parameter used for specifying the accuracy of an instrument.
• The limiting error (or guarantee error) is specified by the manufacturer to define the maximum limit of
the error that may occur in the instrument.
• Suppose the accuracy of a 0-100V voltmeter is specified as 2% of the full-scale range.
• This implies that the error is guaranteed to be within ± 2V for any reading.
• If the voltmeter reads 50V, then also the error is also within ± 2V.
• As a result, the accuracy for this reading will be 2 ×100 = 4% .
• If the overall performance of a 5-measuring system is dependent on the accuracy of several
independent parameters, then the limiting or guarantee error is decided by the absolute error.

12
• The accuracy of a measuring instrument is usually specified by its manufacture as % of full-scale
reading. thus, if the user is not using the instrument in its full scale the error will be higher than the
percentage specified. The limiting error is defined as the maximum deviation in the reading.
• Ex: if we are measuring the value of an unknown resistance element using a Wheatstone bridge
whose known resistors have specified accuracies of 1%, 2% and 3% respectively, then:

Then following the logic given to establish, the absolute

error is computed by taking the positive values only and
the errors will add up; as a result, the limiting error for the
unknown resistor will be 6%.

13
• Incomplete definition (may be systematic or random) — One reason that it is impossible to make exact
measurements is that the measurement is not always clearly defined.
• Ex: if two different people measure the length of the same string, they would probably get different results because each
person may stretch the string with a different tension. The best way to minimize definition errors is to carefully consider and
specify the conditions that could affect the measurement.

• Failure to account for a factor (usually systematic) — The most challenging part of designing an experiment is
trying to control or account for all possible factors except the one independent variable that is being analyzed.
• For instance, you may inadvertently ignore air resistance when measuring free-fall acceleration, or you may fail to account for
the effect of the Earth’s magnetic field when measuring the field near a small magnet.
• The best way to account for these sources of error is to brainstorm with your peers about all the factors that could possibly
affect your result. This brainstorm should be done before beginning the experiment in order to plan and account for the
confounding factors before taking data.
• Sometimes a correction can be applied to a result after taking data to account for an error that was not detected earlier.
14
• Environmental factors (systematic or random) — Be aware of errors introduced by your immediate working
environment. You may need to take account for or protect your experiment from vibrations, drafts, changes in
temperature, and electronic noise or other effects from nearby apparatus.
• Instrument resolution (random) — All instruments have finite precision that limits the ability to resolve small
measurement differences.
• Calibration (systematic) — Whenever possible, the calibration of an instrument should be checked before taking
data. If a calibration standard is not available, the accuracy of the instrument should be checked by comparing with
another instrument that is at least as precise, or by consulting the technical data provided by the manufacturer.
Calibration errors are usually linear (measured as a fraction of the full-scale reading), so that larger values result in
greater absolute errors.
• Physical variations (random) — It is always wise to obtain multiple measurements over the widest range possible.
Doing so often reveals variations that might otherwise go undetected. These variations may call for closer
examination, or they may be combined to find an average value. 15
• Zero offset (systematic) — When making a measurement with a micrometer caliper, electronic balance, or
electrical meter, always check the zero reading first. Re-zero the instrument if possible, or at least measure and
record the zero offset so that readings can be corrected later. It is also a good idea to check the zero reading
throughout the experiment. Failure to zero a device will result in a constant error that is more significant for smaller
measured values than for larger ones.
• Parallax (systematic or random) — Occur whenever there is some distance between the measuring scale and the
indicator used to obtain a measurement. If the observer’s eye is not squarely aligned with the pointer and scale, the
reading may be too high or low.
• Instrument drift (systematic) — Most electronic instruments have readings that drift over time. The amount of drift
is generally not a concern, but occasionally this source of error can be significant.
• Personal errors: come from carelessness, poor technique, or bias on the part of the experimenter. The
experimenter may measure incorrectly, or may use poor technique in taking a measurement, or may introduce a
bias into measurements by expecting (and inadvertently forcing) the results to agree with the expected outcome.
16
• Lag time and hysteresis (systematic) — Some measuring devices require time to reach equilibrium
and taking a measurement before the instrument is stable will result in a measurement that is too high or
low. Ex.: is taking temperature readings with a thermometer that has not reached thermal equilibrium
with its environment. A similar effect is hysteresis where the instrument readings lag and appear to have
a “memory” effect, as data are taken sequentially moving up or down through a range of values.

• Determination Error : It is due to the indefiniteness in final adjustment of measuring apparatus.

• For example, Maxwell Bridge method of measuring inductances, it is difficult to find the differences
in sound of headphones for small change in resistance at the time of final adjustment. The error
varies from person to person.

17
• Error due to Other Factors
• Temperature Variation : Variation in temperature not only changes the values of the parameters
but also brings changes in the reading of the instrument. For a consistent error, the temperature
must be constant.
• Effect of the Time on Instruments : There is a possibility of change in calibration error in the
instrument with time. This may be called ageing of the instrument.
• Effect of External Electrostatic and Magnetic Fields : These electrostatic and magnetic fields
influence the readings of instruments. These effects can be minimized by proper shielding.
• Mechanical Error : Friction between stationary and rotating parts and residual torsion in
suspension wire cause errors in instruments. So, checking should be applied. Generally, these
errors may be checked from time to time.

20/03/22 18
 UNCERTANITY

• Assess experimental procedure including identification of potential

difficulties
– Definition of necessary steps – Gaps
• Advise what procedures need to be put in place for measurement
• Identify instruments and procedures that control accuracy and
– Usually one, or at most a small number, out of the large set of
possibilities
• When experiment cannot meet desired accuracy

20/03/22 19
 Density of gold = 15.5 gm/cm3

Density of alloy = 13.8 gm/cm3

Density g/cm3
• Measurement by Abhishek and Rahul
17
• Abhishek best estimate is 15
• [probable range:13.5 to 16.5 gm/cm3 ] Error Bar 16

Gold
• Rahul best estimate is 13.9
Abhishek 15
• [probable range:13.7 to 14.1 gm/cm ]
3

14
Rahul Alloy

13

20/03/22 20
• Provide the only known basis for deciding whether:
– Data agrees with theory
– Phenomena measured are real
• Provide basis for defining whether a closure check has been achieved
• – Is continuity satisfied (does the same amount of mass go in as goes out?)
• – Is energy conserved?
• Provide an integrated grasp of how to conduct the experiment

20/03/22 21
 ERROR ANALYSIS ON COMMON SENSE BASIS

• Consider the calculation of electric power from

P=E.I,
where,
E and I are measured as : E = 100 V ± 2 V
I = 10 A ± 0.2 A
The nominal value of power is 100 ± 10 = 1000 W.
By taking worst possible variations in voltage and current, we could calculate
Pmax = (100 + 2) (10 + 0.2) = 1040.4 W
Pmin = (100 – 2) (10 – 0.2) = 960.4 W
• The uncertainty in the power is + 4.04 percent or – 3.96 percent.

20/03/22 22
• It is quite unlikely that the power would be in error by these amounts because the voltmeter variation would probably
not correspond with the ammeter variations. When voltmeter reads an extreme ‘high’, there is no reason why the
ammeter must also read an extreme ‘high’ at that particular instant, indeed, this combination is most unlikely.
• The simple calculation applied to the electric-power equation above is a useful way of inspecting experimental data
to determine what error could result in a final calculations.
• However, the test is too severe and should be used only for rough inspection of data.
• It is significant to note, however, that if the results of the experiments appear to be in error by more than the
amounts indicated by the above calculation, then the experimenter had better examine the data more closely.
• In particular, the experimenter should look for certain fixed errors in the instrumentation, which may be eliminated
by applying either theoretical or empirical corrections.

20/03/22 23
 STATISTICAL ANALYSIS OF DATA

• The Experimental Data is obtained in Two Forms of the tests:-

1. Multi sample Test: Repeated measurement of a given quantity are done using different test
conditions such as employing different instruments, Different ways of measurement and by
employing different observers.
2. Single Sample Test: A single measurement for succession of measurement done under identical
conditions excepting for time is known as single sample Test.

20/03/22 24
• Histogram (frequency distribution curve): When the
number of multi sample observations are taken
experimentally, there is a scatter of the data about some
central value. One method presenting test results in the
form of histogram.
• A Histogram is a graphical display of data using bars of
different heights.
• It is similar to a Bar Chart, but a histogram groups numbers
into ranges and you decide what ranges to use.

20/03/22 25
• When a quantity is measured repeatedly it is expected that it will be distributed around the best value according to
some distribution.
• Many times the random errors may be distributed as a normal distribution.
20/03/22 26
• Arithmetic mean: The best approximation Central value
is the same
that can be made of a number of readings of
the same quantity. It is also called Mean
Curve 2: Low precision
value. This mean is computed by summing all
the values and dividing by the number of
Curve 1: Higher precision
measurements.
• Dispersion of the property which denotes the
extent to which the values are dispersed X3 X4
X1 X2
about the central value is termed as
curve 1 has a smaller dispersion than that of the
dispersion. It also known as spread or scatter.
data represented by curve 2.

20/03/22 27
 TERMS DEFINING THE MEASURE OF DISPERSION OF DATA

• Range: The simplest possible measure of dispersion is the range which is the difference between
greatest and least values of data.
• E.g. the range of curve 1 is (X2 - X1) and that of the curve 2 is (X4 - X3).
• Mean Value n is the total number of readings and xi is the value of the individual readings.

• Deviation:The deviation of the individual readings from the mean value

𝑑𝑖 = 𝑥𝑖 −
𝑥𝑚
• Variance or the mean square deviation is used as a measure of the deviation of the set of readings.
for a large number of readings
(n>30), one can safely
approximate the variance as,
The term σ is denoted as standard deviation
20/03/22 28
• Average deviation: Defined as the sum of the absolute values of deviations divided by the number of readings.
The absolute value of deviation is the value without respect to its sign. Average deviation of all the readings is zero.
• Standard deviation: is defined as the square root of the sum of the individual derivations squared, divided by the
number of readings. It is the important term in analysis of random numbers. The term standard deviation is often
used as a measure of uncertainty in a set of measurements. Standard deviation is also used as a measure of
quality of an instrument.
• Standard Deviation of the Mean indicates that the precision can be increased, (i.e., σ (x) reduced) by taking a
greater number of observations. But the improvement is slow due to the n factor.
• Example: Suppose, a measuring instrument produces a random error whose standard deviation is 1%. How many measurements should be
taken and averaged, in order to reduce the standard deviation of the mean to <0.1%?
Solution: n > 100

20/03/22 29
OBSERVATION WIDTH (CM)
1 31.33 n is the total number of readings and xi
2 31.15 is the value of the individual readings.
3 31.26
4 31.02 Mean = 31.19 cm
5 31.20

OBSERVATION WIDTH (CM) Deviation (cm)

1 31.33 + 0.14
2 31.15 - 0.04 Average Deviation = 0.86 cm
3 31.26 + 0.07
4 31.02 - 0.17
5 31.20 + 0.1

30
• An internal estimate can be given by repeat measurements.
• Random error is generally of same size as standard deviation (root mean square deviation) of
measurements
• Mean of repeat measurements is best estimate of true value
• Standard deviation of the mean (random error) is smaller than standard deviation of a single
measurement by

1
Number of measurements
• To increase precision by 10, you need 100 measurements

20/03/22 31
 SIGNIFICANT FIGURES
Experimental values - both measurements and results calculated from these
measurements - should always be reported with only one uncertain digit.

• The right-most digit of any experimental measurement contains an error.

• "Significant figures" are those digits which have significance - which have meaning.
• If you measure the length of a A4 sheet with a mm-ruler and give the result as 279.33 mm, you would be
incorrect.
• The smallest division is a mm, and you can estimate one more digit corresponding to a place between two
divisions.
• You have absolutely no information on hundredths of a mm.
• That is, in this measurement, the tenths place is somewhat uncertain but the hundredths place is completely
uncertain - it has no experimental significance.
• So your reported value should be 279.3 but not 279.33, or 279.

20/03/22 32
• Things were simple before calculators, because at the touch of a button you have eight or nine digits
displayed before your eyes, The calculator makes life difficult because you must decide which of these
digits are uncertain, and you must round off all but one uncertain digit.
• The rules for rounding off are as follows:
• Examine the digits to be discarded.
• If the first digit is a 5 followed by other digits, at least one of which
• If the first digit is larger than a 5, round up. If the first digit is less than a 5, round down. is not zero,
round up.
• If the first digit is 5 followed only by zeros, or by no other digits at all, round up or down to make the
last digit retained even.

20/03/22 33
• Suppose you have the object made up of three separate component parts - a ball, a cylinder and a
plate. You measure the masses by using a precision analytical balance for the ball and the cylinder and
a triple beam balance for the plate.
The masses are .282 gm, 79.545 gm and 422.23 gm.
The total mass is 502.057 gm
The result as written is incorrect - it has two uncertain digits
The mass must be rounded up, and be written as 502.06 gm.
• When calculating an experimental result by adding or subtracting experimental data, round off so only
one uncertain digit remains in the result.

20/03/22 34
• Area of a rectangle. The length, measured with a meter stick is found to be 11.23 cm and the width,
measured with a Vernier calliper, is found to be .332 cm. The area is 3 .72836 cm2
The correct result is 3.73 cm2
Notice that the result has 3 significant digits - the same number as in the width.
This result illustrates the following rule of thumb: When calculating an experimental result by
multiplying or dividing experimental data, round off the result so that it has only as many significant
figures as the data value with the smallest number of significant figures.
• Another quasi-rule often used by practitioners is: Add a significant digit when the first digit is a 1.
Thus 3 x 0.34 = 1.02, not 1.0.

20/03/22 35
• Consider a length 35.9 m, which can also be written as 3590 cm.
But these two numbers have very different meanings.
The first says that the 9 is uncertain while the second implies that only the last 0 is uncertain.
If it is the 9 which is uncertain~ how should we write the result properly in cm? .
The answer, is ,3.59 X 103 cm.
Note that there are now only three significant figures.
• Let us calculate the area of a square, whose side is measured to be
L = 6.71 ± 0.02 cm.
• Let's write this as L = 6.71 x [1 ± 𝜕],
where the bracketed quantity is a mathematical construct, not a measured quantity.
Here, 𝜕 must be equal to ( 0.02 / 6.71 ).
• We get, upon squaring, A = L2= (6.71)2 x [1 ± 𝜕] x [1 ± 𝜕],
where we note that the first and the second 𝜕 are completely correlated. In fact they are one and same.
• Hence, A = 45.0 x [1 ± 2𝜕 + O (𝜕2)].

20/03/22 36
 PROPAGATION OF ERROR

• When measured quantities are combined (added or multiplied together), the rules associated with significant
figures enabled us to make sensible statements about the resulting quantity.
• Consider the area of the rectangle treated above, which was 3.73 cm2.
• The most conservative position we can take about this number is that the area has a value somewhere within the
range bound by 3.700 and 3.800.
• Frequently, the uncertainty is smaller than that indicated by the rules of significant figures
• The question is, "Is there a simple way to track the error propagation?" The answer is yes.
• Propagation of Error in Single Measurement. Let us assume that we made a measurement of an x and ask how
the standard error propagates as different functions of x are computed.
• It is convenient to consider not only the error itself, but also the relative (fractional) error (ux/ x).
𝑃 = 2𝜋𝑁𝑇 ; N is rpm of the engine (2000 rpm) ; T (torque: 70 Nm)

20/03/22 37
• In error theory, we always consider the fractional error to be small compared to 1; i.e., (ux/ x) < < 1.
(Large fractional errors are very unusual in physics. laboratories.)
All expressions that follow are based on this assumption.
5 sheets weigh 1 ounce

postal scale • You multiplied your measured quantity and its

error by 100.
only weighs
1 ounce (28g) • If we use a large scale to weigh a ream of
paper in pounds, and find the weight of a five-
you want to mail a ream of (500 sheets)
sheet letter by division, you would divide both
Under the assumption that all sheets of paper have the same
weight, the weight of a ream is 100 oz, or 6 lb and 4 oz. your result and error by 100.

• You put on enough postage for a 7-lb package. Will your package make it? • To summarize: When a measured quantity is
multiplied (divided) by a constant, the absolute
- That depends on the error of your estimate of the weight.
error is likewise multiplied (divided) by the same
• An error of 0.1 oz., then ream would have an error of 100 x 0.1 oz = 10 oz constant.
your package would weight 6 lb 14 oz, and you are safe.

20/03/22 38
• The length of two scales is given as l1 = 20 cm ± 0.5 cm and l2 = 30 cm ± 0.5 cm, then the final length by adding length of both
scales will be given as 50cm ± 1 cm.
• The mass of a substance is 100 ± 5 g and volume is 200 ± 10 cm3, then the relative error in density will be the sum of percentage

! "#
error in mass that is × 100 = 5% and percentage error in volume that is ×100 = 5%, which is 10%.
"## $##
• Error in case of a measured quantity rose to some power: The relative error in physical quantity raised to a power‘s’ can be
calculated by multiplying ‘s’ with a relative error of the physical quantity.
Suppose, there exist a quantity S = A2, where A is any measured quantity, then relative error in S will be given as:

∆R S∆T
R
= T
The general formula to find relative error in such cases can be written as: Suppose S = AxByCz, , then

∆𝑆 𝑥∆𝐴 𝑦∆𝐵 𝑧∆𝐶

= + +
𝑆 𝐴 𝐵 𝐶
∆& '∆( )∆* $∆+
Example 3 4 2
20/03/22The relative error in S = A B C , will be written as, = + + 39
& ( * +
 Propagation of Errors: More Than One Measurement

• Rule #1: When two measurements are combined by addition or subtraction, use absolute errors, and use the

following - Given 𝑍 = 𝑥 ± 𝑦 𝑤𝑖𝑡ℎ 𝑥, 𝜎𝑥, 𝑦, 𝜎𝑦 then 𝜎𝑧 = 𝜎#$ + 𝜎%$

(In addition of measured quantities, absolute errors are added in quadrature.)

• Rule #2: When two measurements are combined by multiplication or division use relative errors, and use the recipe Given z = a x
y or z = ax/y then

$! $ $#
%
= ( &")2 + '
2 (In a product of measured quantities, relative errors add in quadrature.)

• Rule #3: Keep two most significant digits as errors are propagated. For the final answer keep only the most significant digit in the
absolute error.

20/03/22 40
• The-factor-of-two-rule. Remember that errors usually aren't more precise than ±50%; one significant
figure is all that you can expect in your error estimate.
• Given this, we will now see that, if one source of error ,A, is appreciably larger than another source of
error, B, then B has a negligible effect on the final error.
Suppose A is equal to a 2% error and B is half as large.
Then the final error estimate is
% error =v'(12 +22) =2.24 % = 2 %.
• To one significant figure, the total error is described completely by the larger source A.
• The conclusion is quite general: a successful error analysis finds the largest source of error, whether it
be systematic or random, rather than attempting to add the effects of many small errors.

20/03/22 41
 How to estimate the uncertainty in the estimated variable, if
the uncertainties in the measured parameters are known?

• If the four propagation rules, listed above, become obscure in a very complicated error tracking situation,
you may be forced to evaluate the following general formula:
• The problem can be stated mathematically as, Let y = f(x1, x2 …….xn)
• If the uncertainty (or deviation) in xi is known and is equal to Δxi , (i = 1,2,..n) , what is the overall
uncertainty in the term y?
• Differentiating the above expression, and applying Taylor series expansion, we obtain,
𝛿𝑓 𝛿𝑓 𝛿𝑓
∆y = ∆𝑥& + ∆𝑥$ … . . ∆𝑥'
𝛿 𝑥& 𝛿 𝑥$ 𝛿 𝑥'

• Since Δxi can be either +ve or –ve in sign, the maximum possible error is when all the errors are positive
and occurring simultaneously. The term absolute error is defined as
𝛿𝑓 𝛿𝑓 𝛿𝑓
𝐴𝑏𝑠𝑜𝑙𝑢𝑡𝑒 𝑒𝑟𝑟𝑜𝑟: ∆𝑦 = ∆𝑥& + ∆𝑥$ … … . ∆𝑥'
𝛿 𝑥& 𝛿 𝑥$ 𝛿 𝑥'
42
• But this is a very unlikely phenomenon, as in practice, x1,x2,.....,xn are independent and all errors do not occur
simultaneously.
• Defining cumulative error y in terms of the standard deviation. Squaring equation, we obtain,
$ $
$
𝛿𝑓 $
𝛿𝑓 $
𝛿𝑓 𝛿𝑓
∆𝑦 = ∆𝑥& + ∆𝑥$ + ⋯ … . +2 ∆𝑥& Δ𝑥$ + … … .
𝛿𝑥& 𝛿𝑥$ 𝛿 𝑥& 𝛿𝑥$

• If the variations of x1 , x2 ,..... are independent, positive value of one increment is equally likely to be associated with
the negative value of another increment, so that the some of all the cross-product terms can be taken as zero, in
repeated observations.
• We have already defined variance V as the mean squared error. So, the mean of (Δy)2 for a set of repeated
observations, becomes the variance of y, or

$ $
𝛿𝑓 𝛿𝑓
𝑉(𝑦) = 𝑉(𝑥& ) + 𝑉(𝑥$ ) + ⋯
𝛿𝑥& 𝛿𝑥$ 43
• So, the standard deviation of the variable y can be expressed as:
&
$ $ $
𝛿𝑓 𝛿𝑓
𝜎 y = 𝜎 $ (𝑥& ) + 𝜎 $ (𝑥$ ) + ⋯
𝛿𝑥& 𝛿𝑥$

• When there are more than two measured quantities, you can extend this expressions by adding more terms under
the square-root sign.

44
 EXCERCISE

• A pressure gauge with a measurement range of 0 -10 bar has a quoted inaccuracy of ±1.0% of the full
scale reading.
(a) What is the maximum measurement error expected for this instrument?
(b) What is the likely measurement error expressed as a percentage of the output reading if this
pressure gauge is measuring a pressure of 1 bar?
• The width of a room is measured 10 times by an ultrasonic rule and the following measurements are
obtained (units of meters): 5.381 5.379 5.378 5.382 5.380 5.383 5.379 5.377 5.380 5.381.
The width of the same room is then measured by a calibrated steel tape that gives a reading of 5.374
m, which can be taken as the correct value for the width of the room.
(a) What is the measurement precision of the ultrasonic rule?
(b) What is the maximum measurement inaccuracy of the ultrasonic rule?
20/03/22 45
• A packet of resistors bought in an electronics component shop gives the nominal resistance value as 1000 U and
the manufacturing tolerance as ± 5%. If one resistor is chosen at random from the packet, what is the minimum and
maximum resistance value that this particular resistor is likely to have?
• A particular micrometre is designed to measure dimensions between 50 and 75 mm. What is its
measurement range?

20/03/22 46
• Suppose that the instrument characteristic shown in Figure is that of
a pressure sensor, where the input units are expressed in bars from 1
to 9 bars and the output units are expressed in volts from 1 to 13 V.
(a) What is the maximum nonlinearity expressed as a percentage of
the full scale deflection?
(b) What is the resolution of the sensor as determined by the
instrument characteristic given?

20/03/22 47
• The following resistance values of a platinum resistance thermometer were measured at a range of
temperatures. Determine the measurement sensitivity of the instrument in U/C.

20/03/22 48