Bioinstrumentation and Biosensors

(Bioinstrumentació i Biosensors)

Antoni Ivorra (Coordinator) Laura Becerra

[email protected] [email protected]
51.017, La Nau building 51.017, La Nau building

BIS 2024-2025
M0. Course introduction and basic
concepts on measurement systems

BIS 2024-2025 M0: 2/50

This course mostly deals with theoretical and practical aspects
regarding acquisition of numerical data and signals from the human
body by means of electrodes and electronic sensors. In most cases
the final goal of such acquisition is to produce measurements.

BIS 2024-2025 M0: 3/50

 M0.1. Main definitions
nature
signal conversion +
processing + measurement
computing

measurand sensor (e.g. display)

measurement system

▪ The magnitude from nature we want to measure is known as the measurand

(es. “mensurando”; cat. “mesurand”).

▪ The measurand is associated with a stimulus which is detected by the sensor

(also known as detector).

▪ From the signal provided by the sensor, the measurement system produces a
measured value (also known simply as measurement (es. “medida”; cat.
“mesura”)) which is a representation of the measurand.

▪ Ideally, the measured value will be equal to the true value of the measurand. In
reality, some errors must be expected.
BIS 2024-2025 M0: 4/50
Transducer: any device, such as a microphone or a loudspeaker, that converts
one form of energy into another. The term transducer is only applied when at
least one of the forms of energy (either the input or the output) implies some
sort of information or control purposes (i.e., a signal is implied). A combustion
engine, converting chemical energy into mechanical energy, is not generally
considered to be a transducer.
Some decades ago, it was common to talk about input transducers and output
transducers. Now the term transducer is almost exclusively used in the sense of output
transducer or actuator (meaning that a control signal is converted by the transducer into
a form of actuation energy) or in the sense of intermediate energy conversion systems
(next slide). For instance, loudspeakers and electric motors are transducers or actuators.

Instead of input transducer now is preferred the term sensor.

Sensor: any system – not only human made devices – able to convert one
form of energy into another one in such a way that the second form of energy
contains information regarding the first form of energy (i.e., the stimulus).
Until recently, the term sensor was exclusively used for human made devices in which
the second form of energy was electrical. Now biologists use the term sensor for natural
information systems in organisms. In the context of the course, we will indistinctively use
the terms sensor and electronic sensor.

BIS 2024-2025 M0: 5/50

 (image adapted from “Handbook of modern sensors: physics, designs, and applications” by J. Fraden)
Electronic Sensor (or Detector): any device able to convert one form of
energy (i.e., the stimulus) into an electrical signal in such a way that the
electrical signal contains information regarding the stimulus.

An electronic sensor may contain a chain of transducers in order to obtain an electrical

signal from the stimulus.

electronic sensor

electrical
e1 e2 e3 direct signal
transducer 1 transducer 2
stimulus sensor

A sensor may incorporate several transducers

(ek are signals in different types of energy)

Diccionari termcat de metrologia: http://www.termcat.cat/ca/Diccionaris_En_Linia/192

BIS 2024-2025 M0: 6/50

The flow sensor developed in former lab sessions of the course is an example in which
intermediate transducers are employed before obtaining the electrical signal.

Transducer 1: flow to
pressure flow in flow out
Q Q

+ −
ΔP = kQ

Direct sensor: pressure

to electrical signal
Pressure sensor

In fact, the pressure sensors contains a chain of transducers:

Pressure → Deflection (of a membrane) → Elongation (of conductors on the membrane) → Electrical Resistance

BIS 2024-2025 M0: 7/50

Video: “Bosch Working principle of a pressure sensor”
https://youtu.be/zXIeqeT_FC8
 M0.2. Teaching objectives

The course aims three main teaching objectives:

1- to overview the fundamental features of modern medical measurement

systems and the mathematical, physical and electronic foundations of those
systems,

2- to introduce the principles of design for electronic circuits capable of

safely interfacing electronic sensors and electrodes to digital systems and,

3- to put in practice the above knowledge – and to get some lab and
hardware prototyping skills – by implementing and testing a very simple
medical measurement system.

BIS 2024-2025 M0: 9/50

(This is the area focus of the BIS course)

(figure from “Measurement and instrumentation: theory and application”)

BIS 2024-2025 M0: 10/50

Short syllabus:
M0. Basic concepts on measurement systems.
M1. Circuit theory fundamentals and electrical safety.
M2. Signal conditioning circuits, biopotential amplifiers and analog-to-
digital conversion.
M3. Brief introduction to digital electronics.

Prerequisites:
1. Very basic knowledge of circuit theory ≡ “Bioelectromagnetisme”
2. Undergrad level general physics ≡ “Bioelectromagnetisme”
3. Signal theory basics ≡ “Teoria de Senyals i Sistemes”
4. Matlab programming basics ≡ “Tècniques Computacionals en Biomedicina”

BIS 2024-2025 M0: 11/50

 Hours in classroom
Contents Hours outside the
module LECTURE SEMINAR classroom
LAB
(medium (medium
(small group)
group) group)

M0 2 – – 6
M1 8 10 – 25
M2 6 6 – 16
M3 4 2 – 10
ML – 4 8 18
Total 20 22 8 75 125

BIS 2024-2025 M0: 12/50

 Assessment

Minimum
Weight
Assessment activity grade to Recoverable (in
Assessment methodology and criteria on final
(and typology) pass the July)
grade
course
AL1. Introduction lab test Short multiple-choice test performed at the end of the
0 No 5%
(Written Test) second lab session.

Execution test to assess circuit simulation capabilities.

AL2. SPICE test
It will consist in solving a number of circuit problems by 0 No 10%
(Execution Test)
performing LTspice simulations.

The professor will check the performance and the

AL3. Demonstration of
design (hardware + software) of the implemented
EMG system (Execution 0 No 10%
systems. (The grade will be the same for the members
Test)
of the team.)
AC1. M0-M1 test
Multiple-choice test plus two to six problems. 0 No 15%
(Written test)

AF. Final exam Modules 0 to 3 will be assessed by multiple choice

questions plus three to ten comprehensive problems to 5 over 10 Yes 60%
(Written Test) be solved.

BIS 2024-2025 M0: 13/50

Activity Relevant dates
Test on 28/01/2025
AC1. Feedback test on M0-M1C
Grades publication on 07/02/2025
Test on 12/02/2025
AL1. Intro lab test
Grades publication on 22/02/2025
Test on 27/02/2025
AL2. SPICE test
Grades publication on 08/03/2025
Evaluated on 4-5/03/2025
AL3. Demonstration of EMG system
Grades publication on 15/03/2025
AF4. Final exam (to be announced; most likely on 20/03/2025)

BIS 2024-2025 M0: 14/50

▪ In case a student cannot attend a session in which a feedback test is
performed, his or her grade for that test will be 0. Tests will be repeated for
individual students. Only in exceptional and justified cases, the portion of the
final grade corresponding to the missing test will be transferred to the final
exam.

▪ Punctuality will be strictly enforced. Particularly in lab sessions.

▪ Very important: copies or plagiarism will not be tolerated at all!

(more details in the Learning Plan at , read it!)

BIS 2024-2025 M0: 15/50

 Bibliography

Recommended books for the course:

▪ “Foundations of Analog and Digital Electronic Circuits”, Anant Agarwal and Jeffrey H.
Lang, publisher: Morgan Kaufmann, 2005. Covers most M2 and M3 partially; extensive
collection of problems. [Five copies at the UPF library].
▪ “Principios de Electrónica” (Electronic Principles), Albert Paul Malvino, publisher:
McGraw-Hill, 2007. Covers most M2 and M3 partially; extensive collection of problems.
[Three copies at the UPF library].
▪ “Medical Instrumentation Application and Design”, 4th edition, John G. Webster
(editor), publisher: Wiley, 2009. Covers M1 partially, M2 (safety), M3 (biopotential
amplifiers) and M4 partially; contains some problems. [Three copies at the UPF library].
▪ “Introduction to Biomedical Equipment Technology”, 4th edition, Joseph J. Carr and
John M. Brown, publisher: Prentice Hall, 2000. Covers M1 in detail, M3 and M4
partially; contains some problems. [Two copies at the UPF library].
▪ “Biomedical Device Technology: Principles and Design”, Anthony Y. K. Chan,
publisher: Charles C Thomas Pub Ltd, 2008. Covers M1 partially, M2 (safety), M3
(biopotential amplifiers) and M4; contains a few solved exercises. [Available as
electronic resource at the web site of the UPF library].

BIS 2024-2025 M0: 16/50

 Bibliography

Note: all these listed books focus on different specific topics covered by the course.
Unfortunately, each one of these books either does not cover some topics of the
course or goes much deeper into the topics than what required in the BIS course, or
both. That is, none of them should be considered singly as the reference book for the
course (i.e. textbook).

Additional support books (further reading) are listed in the Learning Plan.

BIS 2024-2025 M0: 17/50

 Bibliography

After the course, if interested in electronics:

“The Art of Electronics”, 3rd edition, Paul Horowitz and Winfield Hill, publisher: Cambridge
University Press, 2015 . [Three copies at the UPF library].

BIS 2024-2025 M0: 18/50

BIS lab project by Daniel Patón, David Romero, Joel Romero and Manuel Obelleiro (March 2020).
 Hardware lab sessions

Four lab sessions, of two hours each, aimed at the development of simple signal acquisition
systems (sensor + electronics + software). Key ideas:
▪ No reports required
▪ Require previous work and study (as indicated in the lab guide)
▪ At the electronics lab 55.028
▪ Punctuality will be strictly enforced
▪ Copies or plagiarism will not be tolerated at all!
▪ Teams of two

The first two lab sessions:

▪ Introduction to the Arduino platform and its interface with MATLAB
▪ Previous study will not be tested directly at the beginning of the sessions but it will be
assessed in a short multiple-choice test at the end of the second lab session (AL1).

The other four lab sessions:

▪ Development of a system (electrodes + electronics + software) capable of displaying
electromyographic signals picked up at the surface of the forearm.
▪ Looser supervision and guidance
▪ Demonstration test and at the end of the fourth session (AL3).

BIS 2024-2025 M0: 20/50

 Video: “An Introduction to the Arduino”
https://www.youtube.com/watch?v=CqrQmQqpHXc&feature=youtu.be
(watch also: https://www.youtube.com/watch?v=nL34zDTPkcs)
 In four weeks from now you will have your first lab session.
Meanwhile:

▪ Choose your labmate (teams of two students).

▪ You should be automatically assigned to either lab group P101,
P102 or P103. If you want to change that assignment, before
Tuesday next week (14/01), send an email to Laura Becerra letting
her know which lab group you and your labmate you would prefer to
be in. Take into account your group in the courses “Differential
equations” and “Biosignals and Biosystems”.
▪ Prepare the lab session some days in advance (you will have to
read some documents and to watch some videos)
▪ If you are repeating the course, you do not need to do the lab work
(AL1, AL2 & AL3) again. Your past lab grade will be considered this
year. But you can do the lab work, and get a new grade, if you want
and if you have a combined grade for AL is below 5.

BIS 2024-2025 M0: 22/50

Let’s continue overviewing some basic concepts in measurement systems…

BIS 2024-2025 M0: 23/50

Why to use sensors?

▪ For measuring and monitoring (medical diagnosis, climate, scientific and

technical purposes, navigation, trade …)

▪ For security and safety applications (burglar alarms, airbags…)

▪ For recording and reproducing nature signals (sound recording, video

recording…)

▪ For human-machine interaction (input devices, computer games…)

▪ Feedback control mechanisms (automotive, industrial, household

appliances…)

BIS 2024-2025 M0: 24/50

 A few words on negative feedback control

“Negative feedback occurs when the result of a

process influences the operation of the process
itself in such a way as to reduce changes.
Negative feedback tends to make a system self-
regulating [...] Negative feedback is widely used in
mechanical and electronic engineering, but it also
occurs naturally within living organisms, and can
be seen in many other fields from chemistry and
economics to social behaviour and the climate.”
(from wikipedia)

Reference value Error signal Q Q

TSET + TSET - TROOM
+ Comparator Heater Room
-

Measured temperature Temperature

TROOM Output temperature
Sensor
(+ signal
conditioning)
BIS 2024-2025 M0: 25/50
 M0.2. Modern measurement systems
General architecture of modern measurement systems:

analog-to-digital
converter

(signal conditioning
electronics)

(figure from “Handbook of modern sensors: physics, designs, and applications” by J. Fraden)

BIS 2024-2025 M0: 26/50

Some electronic devices commercially available integrate most of the functionality of data
acquisition systems and greatly facilitate the implementation of measurement systems.
These devices are particularly adequate for prototyping and experimentation:

▪ Data Acquisition Boards (DAQs), also referred to as

digitizers: these devices are always connected to a
computer and are typically shaped as plug-in boards
(e.g., PCI format). They embed the minimum number of
elements required for real-time digital signal acquisition.
Typically, they also embed digital input/output lines and
digital-to-analogue converters.

▪ Data loggers: these devices are intended to record

data at a specific rate (samples/s) from a sensor or a set
of sensors. Very frequently they embed the sensors.
They contain memory to store the data but, typically, no
means to display the data. Data is downloaded into a
computer after acquisition (typically offline).

▪ Instruments with digital connectivity: many modern

stand-alone instruments embed digital communications
means that can be used to transfer measurements to a
computer. Combining inputs from multiple instruments it
is possible to create a computer virtual instrument with
advanced features.
BIS 2024-2025
 A Data Acquisition
Board (DAQ) contains
these elements
analog-to-digital
converter

(signal conditioning
circuitry)

A microcontroller (μC) such as the one in

Arduino boards contains all these elements

BIS 2024-2025 M0: 28/50

 In the lab sessions we will use the Arduino Uno board as a Data Acquisition Board to get
signals into Matlab

Signal conditioning USB

Sensor 1
electronics (virtual serial port)
Arduino Uno
Board
Sensor 2 Signal conditioning
electronics
power
MATLAB for signal processing
and representation
analog input terminals

Matlab “server”
program running in
Arduino

USB
(virtual serial port)

MATLAB

USB value = a.analogRead(5);

connector % read analog pin 5

(Arduino Uno microcontroller)

digital input/output terminals
BIS 2024-2025 M0: 29/50
 M0.3. Errors and uncertainty
M0.3.1. Absolute and relative errors

sometimes the error is defined as

the absolute value (modulus) of this
difference

𝜀 =error = absolute error = measured value – true value of the measurand

measured value – true value of the measurand ×100%

𝜀𝑟 =relative error =
true value of the measurand

typically, the relative error is expressed in %

BIS 2024-2025 M0: 30/50

 M0.3.2. Accuracy and Precision
Accuracy and Precision
(es. cat. “exactitud”)
Warning, precision is often confused with accuracy!

The accuracy of an instrument is a measure of how close

the output reading of the instrument is to the true value. In
practice, it is more usual to quote the inaccuracy or
measurement uncertainty value rather than the accuracy
value for an instrument. Inaccuracy or measurement
uncertainty is the extent to which a reading might be
wrong and is often quoted as a percentage of the full-scale
(f.s.) reading of an instrument.

Precision is a term that describes an instrument’s degree

of freedom from random errors. If a large number of
readings are taken of the same quantity by a high-
precision instrument, then the spread of readings will be
very small. High precision does not imply anything about
measurement accuracy. A high-precision instrument may
have a low accuracy. Low accuracy measurements from a
high-precision instrument are normally caused by a bias in
the measurements, which is removable by recalibration
(from “Measurement and instrumentation: theory and application”)
 M0.3.3. Systematic vs. random errors
Measurement errors can be divided into two components: random errors and
systematic errors.

Systematic errors are biases in measurement which lead to the situation where the
average of many individual measurements (of the same measurand) differs from the
actual value of the measurand.

Random errors are those caused by lack of repeatability in the output of the
measurement system.

(from “Introduction to Engineering Experimentation”)

If only random errors exist, the average of a sufficiently large population of

measurements will be free from error (the average of random errors is 0, per definition)
BIS 2024-2025 M0: 32/50
The existence of errors implies uncertainty: we cannot be sure
about the correctness of the measurement. However, in most cases,
bounds can be specified for such uncertainty. For each element of a
measurement system, uncertainty bounds can be specified under a
set of restricted conditions

The terms inaccuracy and uncertainty are frequency used as synonymous.

BIS 2024-2025 M0: 33/50

 M0.3.4. Reporting uncertainty
Example (from “Measurement and instrumentation: theory and application”)

A pressure gauge with a measurement range of 0–10 bar has a quoted

inaccuracy of ± 1.0% f.s. (± 1% of full-scale reading).

(a) What is the maximum measurement error expected for this instrument?

(b) What is the maximum measurement error expressed as a percentage of

the output reading if this pressure gauge is measuring a pressure of 1 bar?

(a) The maximum error expected in any measurement reading is 1.0% of the full-scale
reading, which is 10 bar for this particular instrument. Hence, the maximum likely error is
1.0% × 10 bar = 0.1 bar (this is an absolute error: difference between the measurement and
the true value of the measurand)

(b) In this case, the maximum measurement error is a constant value related to the full-scale
reading of the instrument, irrespective of the magnitude of the quantity that the instrument is
actually measuring. The magnitude of the error is 0.1 bar. Thus, when measuring a pressure
of 1 bar, the maximum possible error of 0.1 bar is 10% of the measurement value (this is a
relative error).
Reading will normally be expressed as (1.0 ± 0.1) bar
Measurement value Measurement uncertainty

Measurement result
BIS 2024-2025 M0: 34/50
Another example. 1.5 V measurement with TTi multimeter 1906, 5½ digits:

d stands for “digit”. That “digit” (also known as

LSD; “least significant digit”) indicates the
smallest value that can be represented on the
display (i.e. the resolution). Therefore,
uncertainty is expressed as a fraction of the
reading (variable amount) plus a fixed fraction
of the “span”.
Provided that the indicated conditions are met (in-house calibration, temperature and warm-up time),
a 1.50000 volt reading at the display (2 V range ) has an uncertainty of only ± (0.00018 + 0.00003) V
and should be reported as (1.50000 ± 0.00021) V
Measurement value Uncertainty of the measurement

Measurement result

BIS 2024-2025 M0: 35/50

 M0.3.4. Transfer function
An ideal (nominal) output/input relationship is characterized by the so-called
transfer function (or characteristic). This function ( 𝑦 = 𝑓 𝑥 ) establishes
dependence between the output signal 𝑦 produced by the system and the input
signal 𝑥.

Transfer functions can be defined for parts of measurement system (e.g., the
sensor) or for the whole measurement system.

The function may be stated in the form of a table of values, a graph, or a

mathematical equation.
𝑏 is the slope of the function
Linear function: 𝑦 = 𝑓 𝑥 = 𝑎 + 𝑏𝑥
and it is called sensitivity

For a nonlinear transfer

Logarithmic function: 𝑦 = 𝑓 𝑥 = 𝑎 + 𝑘 ln(𝑥) function, the sensitivity b is not
a fixed value as for the
𝑘𝑥 linear relationship. At any
Exponential function: 𝑦 = 𝑓 𝑥 = 𝑎𝑒 particular input value, 𝑥0 , it can
be defined as:

Power function: 𝑦 = 𝑓 𝑥 = 𝑎0 + 𝑎1 𝑒 𝑘𝑥 𝑑𝑦(𝑥0 )

𝑏 𝑥0 =
𝑑𝑥
BIS 2024-2025 M0: 36/50
 y (output)

Specified accuracy limits

Nominal transfer function

(possible transfer function)

Full-Scale Output (FSO)

y0

y0’

x0 x (input)

Span =Full-Scale Input (FS)

BIS 2024-2025 M0: 37/50
Let’s assume the case of a temperature sensor whose output is in the form of voltage …

voltage

Nominal transfer function

𝑉𝑂 = 𝑓𝑁 𝑇

Actual transfer function

𝑉𝑂 = 𝑓 𝑇
VO

To produce the measurement, the computing or

representation sub-system of the measurement
system assumes the actual transfer function is
equal to the nominal transfer function.
𝑇𝑀 = 𝑓𝑁−1 𝑉𝑂
𝜀

TM T temperature
measured temperature actual temperature

BIS 2024-2025 M0: 38/50

Let’s overview some typical nonidealities of transfer functions…

BIS 2024-2025 M0: 39/50

 M0.3.4.1. Drift (es. cat. “deriva”)

The transfer function may be not stable. Environmental changes (temperature,

(figure from “Handbook of modern sensors: physics, designs, and applications” by J. Fraden)
humidity, light…) and time (i.e. aging) affect the transfer function of measurement
systems in two main ways, known as zero drift (=bias) and sensitivity drift.

BIS 2024-2025 M0: 40/50

 M0.3.4.2. Hysteresis

(figure from “Handbook of modern sensors: physics, designs, and applications” by J. Fraden)
The transfer function may exhibit a “plastic” behavior. Hysteresis is the dependence of a
system not only on its current stimuli but also on its past status.

Typical causes for hysteresis are friction and structural changes in materials.

Transfer function with hysteresis

BIS 2024-2025 M0: 41/50

 M0.3.4.3. Saturation

(figure from “Handbook of modern sensors: physics, designs, and applications” by J. Fraden)
Every system has its operating limits. Even if the system (e.g., a sensor) is
considered linear, at some levels of the input stimuli, its output signal no longer will
be responsive; a further increase in stimulus does not produce a desirable output.
It is said that the system exhibits a span-end nonlinearity or saturation.

Transfer function with saturation

(adapted from “Handbook of modern sensors: physics, designs, and applications” by J. Fraden)

BIS 2024-2025 M0: 42/50

 M0.3.4.4. Limited resolution
The result of the transfer function may be discrete instead of continuous.

Resolution describes the smallest increment of stimulus which can be sensed (under the
assumption of no-noise conditions). In other words, it is lowest limit on the magnitude of
the change in the input measured quantity that produces an observable change in the
system output.

Resolution in a modern measurement system usually determined by the digits of the

display.

In analog instruments, one of the major factors influencing the resolution of the instrument
is how finely its output scale is divided into subdivisions.

As a general rule, we can assume that the

measurement system or the operator rounds to the
last digit and hence the resolution uncertainty will
be ± 0.5 × resolution value.

The resolution of this pressure gauge is 0.1 bar

Assuming that the operator rounds the measurement to
the closest subdivision mark to the needle tip, the
uncertainty due to limited resolution is ± 0.05 bar.
BIS 2024-2025 M0: 43/50
 M0.3.5. Calibration
The distinction between systematic and random errors is also done because systematic
errors sometimes can be eliminated by a calibration process.
(cat. “patrons de mesura”)
Calibration implies measuring a single or a set of well known stimuli (e.g. measurement
standards) and, by comparing the output of the measurement system to the true values,
adjusting or compensating the transfer function of the system so that it minimizes
systematic errors.

Example: a semiconductor sensor for temperature measurement is given to us with

the indication that its transfer function is linear. How can we parameterize such
transfer function by calibration?
We know: 𝑣 = 𝑓 𝑇 = 𝑎 + 𝑏𝑇
Where 𝑣 is the voltage provided by the sensor and 𝑇 is the temperature
To determine constants 𝑎 and 𝑏, such a sensor can be subjected to two known
temperatures (𝑇1 and 𝑇2 ) and two corresponding output voltages (𝑣1 and 𝑣2 ) will be
registered.
𝑣1 = 𝑎 + 𝑏𝑇1 𝑣1 − 𝑣2
𝑏= ; 𝑎 = 𝑣1 − 𝑏𝑇1
𝑣2 = 𝑎 + 𝑏𝑇2 𝑇1 − 𝑇2
Then our measurement system will be able to provide accurate temperature
measurements from voltage readings by computing the inverse transfer function:
𝑣−𝑎
𝑇𝑀𝐸𝐴𝑆 = 𝑓 −1 𝑣 =
𝑏
BIS 2024-2025
Calibration it is very commonly used for zero drift compensation (= offset adjustment = zeroing).

Body weight scales typically include a knob for manual zeroing

(At a known applied weight, 0 kg, the transfer function of the scale is adjusted until the measured
value is 0 kg)
BIS 2024-2025 M0: 45/50
 M0.3.6. Reporting measured numbers

In contrast with exact numbers, measured numbers come with a degree of uncertainty
and this has implications on how to report measured numbers and how to operate with
them.

Exact Numbers: Numbers that are known with complete certainty and have no
uncertainty. They are not subject to measurement error.

Measured Numbers: Numbers obtained through measurements using instruments,

which inherently involve some degree of uncertainty.

Exact numbers Measured Numbers

Measurement using
Source Counting or definition
instruments
Always present (limited
Uncertainty None
accuracy)
Examples 10 chairs 2.34 meters
Is it relevant the number of
No Yes
reported digits?

BIS 2024-2025 M0: 46/50

 Exact numbers vs. measured numbers

Video: “Unit Conversion & Significant Figures Crash Course Chemistry”

https://youtu.be/hQpQ0hxVNTg?si=qcxJ9YHX0gNfCx2H
 How many digits to report?

1. De facto rule (not formally stated): even if no uncertainty value is reported, it is

implicitly assumed that the number of significant digits of a reported measured
magnitude conveys information on its uncertainty: it must be assumed that the reported
value has been rounded.
For example: if it is reported that the length of an object is 1.2 mm, do not assume its true length is
1.200...…00 mm, assume it is between (1.20 ± 0.05) mm

Significant digits: (from http://www.physics.uoguelph.ca/tutorials/sig_fig/SIG_dig.htm *)

Non-zero digits are always significant. Thus, 22 has two significant digits, and 22.3 has three
significant digits.

With zeroes, the situation is more complicated:

a. Zeroes placed before other digits are not significant; 0.046 has two significant digits.
b. Zeroes placed between other digits are always significant; 4009 kg has four significant digits.
c. Zeroes placed after other digits but behind a decimal point are significant; 7.90 has three significant
digits.
d. Zeroes at the end of a number are significant only if they are behind a decimal point as in (c).
Otherwise, it is impossible to tell if they are significant. For example, in the number 8200, it is not
clear if the zeroes are significant or not. The number of significant digits in 8200 is at least two,
but could be three or four. To avoid uncertainty, use scientific notation to place significant zeroes
behind a decimal point:
8.200 × 103 has four significant digits
8.2 × 103 has two significant digits

BIS 2024-2025 M0: 48/50

 How many significant digits to report? (cont.)

2. By performing mathematical operations you are not going to reduce uncertainty: do

not report derived magnitudes with higher resolution than input data. Use the calculator
with common sense.
If your input data is 1.2 mm and you must perform an inversion, do not report 0.8333333 mm-1.,
report 0.83 mm.

However, for avoiding round-off errors, retain as many digits as possible for performing subsequent
calculations (“keep at least one more significant digit than needed in your final answer”, from *).

2.1 The result of a multiplication or division must not be reported with more significant digits
than the number of significant digits of the input factor with less significant digits.

2.2 In summations or subtractions, the number of reported decimal places (not significant
digits) should be the same as the least number of decimal places in any of the numbers being
added or subtracted.

1.1 kg + 20 g → (1.1 + 0.020) kg = (1.120) kg  report 1.1 kg (note that the 20 g become
insignificant)

1.040 + 0.2133 + 0.0134 = 1.2667  report 1.267

Warning, it would be incorrect to do 1.040 + 0.2133 + 0.0134  1.040 + 0.213 + 0.013 = 1.266

BIS 2024-2025 M0: 49/50

 How many significant digits to report? (cont.)

3. Uncertainties and tolerances should be generally quoted to two significant digits at

most. Round uncertainties up (ceil operator).

4. Reported magnitudes must be rounded to be consistent with their uncertainties.

For example, if R=10.05763 Ω with an uncertainty of 27 mΩ, R must be rounded to 10.058 Ω, that is,
the reported value must be R = (10.058 ± 0.027) Ω

(from http://www.physics.uoguelph.ca/tutorials/sig_fig/SIG_dig.htm )

The Two Greatest Sins Regarding Significant Digits:

1. Writing more digits in an answer (intermediate or final) than justified by the number of
digits in the data.

2. Rounding-off, say, to two digits in an intermediate answer, and then writing three digits
in the final answer.

BIS 2024-2025 M0: 50/50