Mathematical Physics: Error Analysis
❖ CSIR-NET PYQ’s
❖
1. The radioactive decay of a certain material satisfies Poisson statistics with a mean rate
of 𝜆 per second, what should be the minimum duration of counting (in seconds) so that
the relative error is less than 1% ? [CSIR JUNE 2012]
(a) 100/𝜆 (b) 104 /𝜆2 (c) 104 /𝜆 (d) 1/𝜆
2. The approximation cos 𝜃 ≈ 1 is valid up to 3 decimal places as long as |𝜃| is less than:
(take 180∘ ) 𝜋 = 57.29∘ ). [CSIR JUNE 2013]
(a) 1.28∘ (b) 1.81∘ (c) 3.28∘ (d) 4.01∘
𝜋𝑃𝑎4
3. The viscosity 𝜂 of a liquid is given by Poiseuille's formula 𝜂 = . Assume that 𝑙 and 𝑉
8𝐼𝑉
can be measured very accurately, but the pressure 𝑃 has an rms error of 1% and the
radius 𝑎 has an independent rms error of 3%. The rms error of the viscosity is closest to
[CSIR JUNE 2015]
(a) 2% (b) 4% (c) 12% (d) 13%
4. Two data sets A and B consist of 60 and 10 readings of a voltage measured using
voltmeters of resolution of 1mV and 0.5mV respectively. The uncertainty in the mean
voltage obtained from the data sets 𝐴 and 𝐵 are 𝑈𝐴 and 𝑈𝐵 , respectively. If the
uncertainty of the mean of the combined data sets is 𝑈𝐴𝐵 , then which of the following
statements is correct? [CSIR DEC 2015]
(a) UAB < UA and UAB > UB (b) UAB < UA and UAB < UB
(c) UAB > UA and UAB < UB (d) UAB > UA and UAB > UB
5. The decay constants 𝑓𝑝 of the heavy pseudoscalar mesons, in the heavy quark limit,
168
are related to their masses 𝑚𝑝 by the relation
𝑎
𝑓𝑃 =
√𝑚𝑝
where 𝑎 is an empirical parameter to be determined. The values
𝑚𝑝 = 6400 ± 160MeV and 𝑓𝑝 = 180 ± 15MeV correspond to uncorrelated
measurement of a meson. The error on the estimate of 𝑎 is [CSIR JUNE 2016]
(a) 175(MeV)3/2 (b) 900(MeV)3/2
(c) 1200(MeV)3/2 (d) 2400(MeV)3/2
6. The resistance of a sample is measured as a function of temperature, and the data are
shown below.
𝑇 ( ∘ C) 2 4 6 8
𝑅(Ω) 90 105 110 115
The slope of 𝑅 vs 𝑇 graphs, using a linear least-squares fit to the data, will be
[CSIR DEC 2016]
(a) 6Ω/ ∘ C (b) 4Ω ∘ C (c) 2Ω/ ∘ C (d) 8Ω/ ∘ C
7. The experimentally measured values of the variables 𝑥 and 𝑦 are 2.00 ± 0.05 and
3.00 ± 0.02, respectively. What is the error in the calculated value of 𝑧 = 3𝑦 − 2𝑥 from
the measurements? [CSIR JUNE 2017]
(a) 0.12 (b) 0.05 (c) 0.03 (d) 0.07
8. Both the data points and a linear fit to the current vs voltage of a resistor are shown in
the graph below.
If the error in the slope is 1.255 × 10−3 Ω−1 , then the value of resistance estimated from
169
the graph is [CSIR JUNE 2017]
(a) (0.04 ± 0.8)Ω (b) (25.0 ± 0.8)Ω
(c) (25 ± 1.25)Ω (d) (25 ± 0.0125)Ω
9. Two physical quantities 𝑇 and 𝑀 are related by the equation
2𝜋 𝑀 + 𝑏
𝑇= √ ,
𝑎 2
where 𝑎 and 𝑏 are constant parameters. The variation of 𝑇 as a function of 𝑀 was
recorded in an experiment to determine the value of 𝑎 graphically. Let 𝑚 be the slope of
the straight line when 𝑇 2 is plotted vs. 𝑀, and 𝛿𝑚 be the uncertainty in determining it.
The uncertainty in determining 𝑎 is [CSIR DEC 2017]
𝑎 𝛿𝑚 𝛿𝑚 𝑏 𝛿𝑚 2𝜋 𝛿𝑚
(a) ( ) (b) 𝑎 ( ) (c) ( ) (d) ( )
2 𝑚 𝑚 2𝑎 𝑚 𝑎 𝑚
10. In an experiment to measure the acceleration due to gravity 𝑔 using a simple pendulum,
the length and time period of the pendulum are measured to three significant figures.
The mean value of 𝑔 and the uncertainty 𝛿𝑔 of the measurements are then estimated
using a calculator from a large number of measurements and found to be 9.82147 m/s
and 0.02357 m/s2 , respectively. Which of the following is the most accurate way of
presenting the experimentally determined value of 𝑔 ? [CSIR JUNE 2019]
(a) 9.82 ± 0.02 m/s2 (b) 9.8215 ± 0.02 m/s2
(c) 9.82147 ± 0.02357 m/s2 (d) 9.82 ± 0.02357 m/s2
170
11. A student measures the displacement 𝑥 from the equilibrium of a stretched spring and
reports it be 100𝜇m with a 1% error. The spring constant 𝑘 is known to be 10 N/m with
1
0.5% error. The percentage error in the estimate of the potential energy 𝑉 = 𝑘𝑥 2 is
2
[CSIR DEC 2019]
(a) 0.8% (b) 2.5% (c) 1.5% (d) 3.0%
12. In an experiment, the velocity of a non-relativistic neutron is determined by measuring
the time (∼ 50 ns) it takes to travel from the source to the detector kept at a distance 𝐿.
Assume that the error in the measurement of 𝐿 is negligibly small. If we want to
𝛿𝑇
estimate the kinetic energy 𝑇 of the neutron to within 5% accuracy i.e., | | ≤ 0.05, the
𝑇
maximum permissible error |𝛿𝑇| in measuring the time of flight is nearest to
[CSIR JUNE 2021]
(a) 1.75 ns (b) 0.75 ns (c) 2.25 ns (d) 1.25 ns
13. Four students (S1 , S2 , S3 and S4 ) make multiple measurements on the length of a table.
The binned data are plotted as histograms in the following figures
(a)S2 (b)S4 (c)S1 (d)S
171
14. A DC motor is used to lift a mass M to a height H from the ground. The electric energy
delivered to the motor is VIt, where 𝑉 is the applied voltage, 𝐼 is the current and 𝑡 the
time for which the motor runs. The efficiency e of the motor is the ratio between the
work done by the motor and the energy delivered to it. If 𝑀 = 2.0 ± 0.02 kg, ℎ = 1.00 ±
0.01 m, 𝑉 = 10.0 ± 0.1 V, 𝐼 = 2.00 ± 0.02 A and 𝑡 = 300 ± 15 s, then the fractional error
|𝛿𝑒/𝑒| in the efficiency of the motor is closest to [CSIR JUNE2023]
(a) 0.05 (b) 0.09 (c) 0.12 (d) 0.15
15. In the measurement of a radioactive sample, the measured counts with and without the
sample for equal time intervals are 𝐶 = 500 and 𝐵 = 100, respectively. The errors in
the measurements of 𝐶 and 𝐵 are |Δ𝐶| = 20 and |Δ𝐵| = 10, respectively. The net error
|Δ𝑌| in the measured counts from the sample 𝑌 = 𝐶 − 𝐵, is closest to
[CSIR DEC 2023]
(a)22 (b)10 (c)30 (d)43
16. A set of 100 data points yields an average 𝑥‾ = 9 and a standard deviation 𝜎𝑥 = 4. The
error in the estimated mean is closest to [CSIR DEC 2024]
(a)3.0 (b)0.4 (c)4.0 (d)0.3
❖ JEST PYQ’s
1. The length and radius of a perfect cylinder are each measured with an RMS error of 1%.
The RMS error on the inferred volume of the cylinder is roughly [JEST 2012]
(a) 1.7% (b) 3.3% (c) 0.5% (d) 1%
2. An astrophysical observation measured the mass of a star as (12.41 ± 1.12)𝑀◻ , where
𝑀◻ is the mass of the Sun. Another independent observation measured the mass of the
same star as (8.40 ± Δ)𝑀◻ . Assuming the errors to have Gaussian distributions, one
172
concluded that the two measurements differed by 3 standard deviations. The value of Δ
was approximately [JEST 2021]
(a) 0.22 (b) 0.73 (c) 1.04 (d) 1.55
❖ TIFR PYQ’s
1. A detector is used to count the number of 𝛾 rays emitted by a radioactive source. If the
number of counts recorded in exactly 20 seconds is 10000, the error in the counting
rate per second is [TIFR 2010]
(a) ±5.0 (b) ±22.4 (c) ±44.7 (d) ±220.0
2. A liquid is flowing through a capillary tube of inner radius 𝑟 under the influence of an
external pressure 𝑃. The uncertainties in the measurements of 𝑃 and 𝑟 are found to be
2% and 1%, respectively. The uncertainty in the flow of liquid per second is
[TIFR 2017]
(a) 4.47% (b) 2.23% (c) 2.83% (d) 3.61%
3. In an experiment that measures the resistivity 𝜌 of a substance it was observed that 𝜌
varies with temperature 𝑇 and a parameter Δ, as
𝜌 = 𝜌0 𝑒 Δ/𝑇
where 𝜌0 is a constant. In one measurement, made at 𝑇 = 100 K and Δ = 50, the
percentage error in Δ is found to be 2% while the percentage error in 𝑇 was 3%. What
was the approximate percentage error for the resistivity 𝜌 ? [TIFR 2021]
(a) 18% (b) 3.6% (c) 9% (d) 1.8%
4. In an archery contest, the aim is to shoot arrows at the center of a board. Three archers,
Amar, Akbar and Anthony each shot 5 arrows at the board. The locations of their arrow
hits are shown in the figures with red stars. Which of the following statements are true?
[TIFR 2021]
173
(a) Amar has more accuracy than Anthony
(b) Amar has more precision than Akbar
(c) Akbar has more accuracy than Anthony
(d) Akbar has more precision than Anthony
5. A very sensitive spring balance with spring constant 𝑘 = 2 × 108 Nm−1 is operating at a
temperature of 300 K. The thermal fluctuations can lead to an error in the
measurement of mass. If you are trying to measure a mass of 1mg, the relative error in
the measurement is closest to [TIFR 2021]
(a) 0.01% (b) 10.0% (c) 20.0% (d) 0.9%
6. Two students A and B, measure the time period of a simple pendulum in the laboratory
using the same stopwatch but following two different methods.
• Student A measures the time taken for one oscillation and repeats it for 𝑁A number of
times and finds the average.
• Student B, on the other hand, measures the time taken for 𝑁B number of oscillations
and then computes the period.
Given that 𝑁A , 𝑁B ≫ 1, to ensure that both students measure the time period with the
same uncertainty, the relation between 𝑁A and 𝑁B must be [TIFR 2022]
(a) 𝑁A = √𝑁B (b) 𝑁A = 𝑁B2 (c) 𝑁A = 𝑁B (d) ln 2𝑁A = 𝑁B
7. In a standardized entrance exam, the passing rates for the past 10 years are tabulated
below.
174
If 1000 candidates appear for the exam every year, the probability that more than 250
students will pass the exam this year is about [TIFR 2022]
(a) 20% (b) 6% (c) 25% (d) 0.1%
8. A spectrographic method to search for exoplanets is by measuring its velocity along the
line of sight, using the Doppler shift in the spectrum. If a star of mass 𝑀 and a planet of
mass 𝑚 are moving around their common centre of mass, this component of velocity
will vary periodically with an amplitude.
2𝜋𝐺𝑁 1/3 𝑚
𝐴=( )
𝑇 𝑀2/3
For a particular planet-star system, if the time period is 𝑇 = (12 ± 0.3) years, and 𝐴 and
𝑀 are measured with an accuracy of 3% each, then the error in the measurement of the
mass 𝑚 is [TIFR 2022]
(a) 3.7% (b) 8.5% (c) 5.8% (d) 6.3%
9. A faint star is known to emit light of a given frequency at an average rate of 10 photons
per minute. An astronomer plans to measure this rate using a photon-counting
detector. If she wants to measure the rate to an accuracy of 5%, approximately how
long should be the exposure time? [TIFR 2023]
(a) 40 minutes (b) 1 hour (c) 20 minutes (d) 10 minutes
10. A particular counting system has an average background rate of 50 counts/min. A
decaying radioisotope source was introduced and the total 168 counts were measured
in one minute. After a delay of 24 hrs, the system measured total 91 counts in one
minute. If these measurements were used for determining the half-life (𝜏) of the source
and if the average background rate, and the time have no errors, the % error
(100 × 𝜎𝜏 /𝜏) in the calculated half-life value due to counting statistics would be:
[TIFR 2024]
175
(a) 24.3% (b) 21.2% (c) 25.7% (d) 18.2%
11. Consider a universe that always expands with a scale factor 𝑎 that increases with time
as 𝑎(𝑡) = 𝐶𝑡 2/3 where 𝐶 is a constant. Its expansion rate at time 𝑡 is defined by the
Hubble parameter
1 𝑑𝑎(𝑡)
𝐻(𝑡) =
𝑎(𝑡) 𝑑𝑡
The current value of 𝐻(𝑡) in the universe is given by 𝐻0 = 975 km s −1 Mpc −1 where
1Mpc = 3.1 × 1022 m. What is the approximate age of this universe? [TIFR 2024]
(a) 109 years (b) 107 years (c) 1011 years (d) 1013 years
12. For a given measurement of particles in a counter, a 10-minute data collection resulted
in a statistical uncertainty of 2.5%. How much additional time must be allocated to
reduce the statistical uncertainty to 0.5% ? [TIFR 2025]
(a) 240 minutes (b) 40 minutes (c) 250 minutes (d) 50 minutes
13. Two students perform a counting experiment independently. Student A measures the
counts for 1-minute intervals each and repeats the measurement five times. The
obtained counts are given below.
Measurement turn Counts
1 25
2 35
3 30
4 23
5 27
This student then takes the mean of these counts and reports the count rate
(counts/min). The second student (B) makes one measurement for five minutes. She
176
measures 145 counts and reports the count rate (counts /min ). If the clock used for all
these measurements is accurate up to 0.1 minutes, and there are no other sources of
uncertainties, we can conclude that:
(a) The count rate reported by student A will have a larger uncertainty than that
reported by student B.
(b) The count rate reported by student B will have a larger uncertainty than that
reported by student A.
(c) The reported uncertainty in both results would be identical
(d) Nothing may be concluded about the relative uncertainties between A and B
Answer Key
CSIR-NET PYQ
1. c 2. b 3. c 4. a 5. c 6. b 7. a 8. b 9. a 10. a
11. b 12. d 13. c 14. a 15. a 16. b
JEST PYQ
1. a 2. b
TIFR PYQ
1. 2. a 3. d 4. d 5. d 6. b 7. b 8. a 9. a 10. a
11. a 12. a 13. a
177