AERO61021 Experimental Methods
AERO61021 Overview
- 15 credits
- The University of Manchester
- Semester 1 2026/27
- Postgraduate, FHEQ level 7
- Lectures, laboratories and project
AERO61021 Experimental Methods is the Semester 1 module on how engineers design, run, analyse and report their own experiments.
- Assessment split Other and Report each carry 50% of the module result.
- Core habit Record raw voltages and calibration details before converting to physical units.
- Dynamic signals A first-order sensor both attenuates and delays an oscillating input.
- Sampling rule Sample at least twice the highest signal frequency, and far faster for amplitude.
What you get in the full AERO61021 Study & Assessment Guide
- 10 chapters, 47 pages — every assessed topic
- 28 original practice questions, most with model answers
- 10 worked examples, solved the way AERO61021 marks
- 20 computed figures, redrawn from the course’s own numbers
- Before-you-submit checklist for the Other (50%) — what markers reward, what costs marks
Free on this page: assessment table, chapter map, one worked example, key terms, FAQ. The rest is in the full study & assessment guide.
How AERO61021 is assessed
| Component | Weight | Format |
|---|---|---|
| Other | 50% | The official course-unit page names this method Other. Its feedback notes describe quizzes: common mistakes are covered in class the following week and solutions are released online the day after. The study-hours table also lists eAssessment time. |
| Report | 50% | The official course-unit page lists a Report. Its feedback notes describe formative drop-in sessions through the semester and group and cohort-wide feedback after submission. The laboratory handout links the LabVIEW work to a group project later in the semester. |
Both weights come from the University of Manchester's official course-unit page for Experimental Methods, published under the cross-listed code AERO62021 that shares this module's teaching site; the teaching materials themselves give no weights or due dates, so confirm deadlines and task rules on Canvas.
What other students are asking about AERO61021
- Where can I find lecture notes or study guides for AERO61021 at manchester?
- The AERO61021 study guide in AskSia Library covers Experimental Methods in 10 chapters, with the assessment table, a chapter-by-chapter topic map, key terms and a worked example, free to read on this page.
- Is there a final exam in AERO61021?
- No final exam is listed in the offering this AskSia Library guide is built from; AERO61021 is assessed by Other 50%, Report 50%.
- What does AERO61021 cover on sensor characteristics and measurement terms?
- In AskSia Library's AERO61021 guide, the Sensor Characteristics and Measurement Terms chapter explains range and full scale, sensitivity and ratiometric output, percentage non-linearity, offset and the hysteresis error at mid-range.
- What does AERO61021 cover on data types and measurement references?
- In AskSia Library's AERO61021 guide, the Data Types and Measurement References chapter explains qualitative and quantitative data, absolute, gauge and differential references, and point measurements against field techniques such as pressure-sensitive paint.
- What does AERO61021 cover on calibration standards and resolution?
- In AskSia Library's AERO61021 guide, the Calibration Standards and Resolution chapter explains calibration curves, traceable chains to SI base units, secondary laboratory standards, calibration records, resolution against accuracy and spatial and temporal resolution.
What AERO61021 covers
Ten chapters follow a measurement from the reason for testing, through sensor behaviour, references, calibration, dynamic response and sampling, to sensor choice, number formats, LabVIEW and the DAQmx acquisition chain.
Why Engineers Experiment
Analysis, simulation and experimentation compared, verification against validation, the cost of model confidence and the stages of a typical experiment.Why Engineers Experiment sets out where physical testing fits among the three ways engineers gather information: analysis, modelling and simulation, and experimentation.
- Analysis, simulation and experimentation as information routes
- The stages of a typical experiment from rig to conclusion
- Verification versus validation of models
- Why confidence in a model costs more near certainty
- Testing at the edges of an operating envelope
- Finding prior experiments, handbooks and data sources
Sensor Characteristics and Measurement Terms
Range and full scale, sensitivity and ratiometric output, percentage non-linearity, offset and the hysteresis error at mid-range.Sensor Characteristics and Measurement Terms builds the vocabulary used on every sensor data sheet and turns it into numbers you can calculate.
- Range and full-scale range of sensors and converters
- Sensitivity as the gradient of the response
- Ratiometric ratings and excitation voltage
- Percentage non-linearity from calibration residuals
- Offset and its dependence on temperature
- Hysteresis error from rising and falling sweeps
Data Types and Measurement References
Qualitative and quantitative data, absolute, gauge and differential references, and point measurements against field techniques such as pressure-sensitive paint.Data Types and Measurement References covers three distinctions that decide how a measurement should be recorded and interpreted.
- Qualitative and quantitative information
- Discrete and continuous quantitative data
- Extracting numbers from images such as Schlieren photographs
- Absolute, gauge and differential reference conditions
- Converting gauge readings with the measured ambient pressure
- Point measurements and spatial coverage
Calibration Standards and Resolution
Calibration curves, traceable chains to SI base units, secondary laboratory standards, calibration records, resolution against accuracy and spatial and temporal resolution.Calibration Standards and Resolution explains how an instrument reading becomes a trustworthy physical value.
- Calibration as a fitted relationship between reading and quantity
- Linear and non-linear calibration laws for potentiometers
- Traceable calibration chains and secondary laboratory standards
- The seven SI base units and national measurement bodies
- Calibration records and storing raw voltages
- Resolution versus accuracy and percentage of full-scale output
Sensor Response Time and Dynamic Signals
Mean and fluctuating components, first-order step response, time constant from a rise time, magnitude ratio and phase lag under oscillating input.Sensor Response Time and Dynamic Signals deals with measurements that change while you are taking them.
- Periodic, quasi-periodic and transient signals
- Mean and fluctuating components of a signal
- First-order exponential response to a step input
- Recovering a time constant from a rise time
- Magnitude ratio for a sinusoidal input
- Phase angle and time lag between input and output
Sampling Rate and Aliasing
The Nyquist-Shannon criterion, the aliasing regimes below twice the signal frequency and why amplitude needs a much faster sample rate.Sampling Rate and Aliasing covers the decision that most often ruins a dynamic dataset before analysis even begins: how fast to sample.
- What aliasing does to a sampled record
- The Nyquist and Shannon sampling results
- The criterion that sample rate must be at least twice the signal frequency
- Positive, destructive and negative aliasing regimes
- Oversampling to recover peak-to-peak amplitude
- Per-channel rates on shared acquisition hardware
Choosing and Powering Sensors
Sensor range and dead weight, pressure sensor types, ratiometric output, amplification, excitation, piezo devices and current, voltage or digital outputs.Choosing and Powering Sensors explains how an experimenter moves from a physical quantity to a device that can actually be wired, powered and read.
- Transducers, transmitters and digital sensors
- Monitoring duty versus experimental duty
- Choosing a measurement range with dead weight included
- Offset voltage outputs and transfer lines
- Amplification of millivolt and microvolt signals
- Excitation, controllers and ratiometric outputs
Binary Number Formats in Data Acquisition
Unsigned and signed integers, two's complement, overflow wrap-around and a floating-point value misread as an integer.Binary Number Formats in Data Acquisition explains why the number a computer displays can differ from the value a sensor produced. Acquisition hardware stores every reading as bits, and software applies a rule to interpret them.
- Unsigned integers from bit weights
- Sign-magnitude representation and its two zeros
- Two's complement bytes and their range
- Overflow and wrap-around in signed formats
- Floating-point words misread as integers
- Choosing a data type for counters and averages
LabVIEW Programming Structures
Front panel and block diagram, data types and wire colours, loops, tunnels, shift registers, arrays, clusters, case structures and loop timing.LabVIEW Programming Structures covers the graphical language used for data acquisition, control and the group project in this module.
- Virtual instruments, subVIs, front panels and block diagrams
- Controls, indicators, constants and wire colours
- Dataflow execution and explicit type conversion
- For Loops, While Loops and tunnel behaviour
- Shift registers and why they must be initialised
- Arrays, auto-indexing and clusters
Acquiring and Saving Data with DAQmx
Virtual channels, single-ended and differential wiring, sample timing, reading blocks, charts against graphs, calibration coefficients and saving to CSV or TDMS.Acquiring and Saving Data with DAQmx follows a voltage from a sensor on a breadboard into a LabVIEW array and out to a file, which is the core task of the hands-on lab and of the later group project.
- The mioDAQ multifunction device and the DAQmx library
- Create Virtual Channel, timing, read and closing the task
- Terminal modes: differential, RSE, NRSE and pseudodifferential
- Sample rate, samples per channel and block duration
- Single-ended and differential potentiometer wiring
- Linear calibration coefficients and their units
The University describes it as the science of experimentation from planning through to data delivery, with an introduction to data reduction and examples such as wind tunnel testing and structural health monitoring.
The teaching begins with why experiments are still needed when analysis and simulation are cheap: models break down at the edges of an operating envelope, and confidence in a model becomes very expensive as it approaches certainty.
It then builds the vocabulary every measurement depends on: range, sensitivity, non-linearity, offset, hysteresis, resolution and accuracy, absolute, gauge and differential references, and the contrast between point sensors and field techniques such as pressure-sensitive paint, digital image correlation and particle image velocimetry.
The second strand treats a measurement as a chain.
A sensor converts a physical quantity into a voltage, current or digital value; calibration relates that output back to the quantity you actually want; the acquisition hardware samples it at a chosen rate; and the computer stores it in a number format that has limits.
Each link has a characteristic failure: a load cell sized without its dead weight, a first-order sensor that attenuates and delays an oscillating input, a sample rate below twice the signal frequency that aliases, or a signed 8-bit value that wraps round to a negative number.
The notes quantify these with simple relations: percentage non-linearity, the hysteresis error at mid-range, a time constant from a quoted rise time, the magnitude ratio and phase lag of a first-order system, and resolution as a percentage of full-scale output.
The practical strand uses LabVIEW and the DAQmx library with a National Instruments mioDAQ.
Lectures cover data types, loops, shift registers, arrays, clusters, case structures and timing, and the first hands-on laboratory wires a potentiometer into an analogue input, reads it, plots it, calibrates the voltage into an angle and saves the data. The laboratory handout states that this task is highly representative of the group project later in the semester.
The official course-unit page lists two assessment methods, Other at 50% and Report at 50%, and its feedback notes describe quizzes and a report supported by drop-in sessions.
Worked example · free
Find a time constant from a rise time, then predict lag
- 2Rearranging the first-order step response with the reading at 90% of the input gives t/τ = −ln(1 − 0.90) = ln 10 = 2.3026, so τ = 46 / 2.3026 = 19.98 s.
- 1The angular frequency is ω = 2πf = 2π × 0.02 = 0.1257 rad/s, so τω = 19.98 × 0.1257 = 2.510.
- 1The magnitude ratio is M = 1 / √(1 + 2.510²) = 1 / 2.702 = 0.370, so the output swings at about 37% of the true amplitude.
- 2The phase lag is φ = tan⁻¹(2.510) = 68.3° = 1.192 rad, and the time lag is ΔT = φ / ω = 1.192 / 0.1257 = 9.48 s.
Key terms
- Validation
- Evidence that a model is accurate enough for its intended use inside the range of conditions it claims to cover, usually gathered by comparing it with experimental data at matching conditions.
- Verification
- Checking that a computerised or analytical model has been implemented correctly and solves the problem it was intended to solve.
- Transducer
- A device that turns a physical input, such as a force or pressure, into a different kind of output, usually a voltage.
- Full-scale output
- The span of a sensor's measurement range, used as the basis when resolution or error is quoted as a percentage.
- Percentage non-linearity
- The largest absolute residual from a straight-line fit expressed as a percentage of the full-scale range.
- Hysteresis
- A response that depends on whether the input is rising or falling, so the two paths form a loop.
- Traceable calibration
- An unbroken chain of calibrations, each with stated uncertainty, leading back to a national or international standard.
- Time constant
- The characteristic time of a first-order sensor that sets how quickly its output approaches a step change in input.
- Aliasing
- An under-sampling effect in which recorded samples appear to have a different frequency content from the real signal.
- Ratiometric output
- An output voltage that is proportional to the excitation voltage as well as to the measured quantity.
- Two's complement
- The signed integer scheme in which the leading bit carries a negative weight, giving an 8-bit range of minus 128 to 127.
- Shift register
- A LabVIEW loop terminal pair that passes a value from one iteration into the next.
- TDMS file
- A National Instruments binary file format that can hold a whole experiment with several runs in one file.
AERO61021 FAQ
How hard is this module?
The difficulty comes from breadth rather than any single technique. In one semester you move from measurement theory and sensor physics to graphical programming and real hardware. The relations are short, but you must choose the right one: rise time to time constant, residual to non-linearity, sample rate to frequency limit.
The University lists 150 study hours, of which 50 are project supervision, so steady practical work matters.
How is the module assessed?
The official course-unit page gives two methods of equal weight: Other at 50% and Report at 50%. Its feedback notes describe two quizzes, with common mistakes reviewed in class the following week and solutions online the next day, plus formative drop-in sessions for the report. The page does not state hurdle rules or dates, so read each task brief on Canvas.
Is there a group project in this module?
Yes. The module's stated aims include developing technical group working skills through a group project, and the first laboratory handout says its tasks are highly representative of the group project later in the semester. That project uses the mioDAQ device with LabVIEW, so the laboratory skills of wiring, reading, calibrating and saving data carry straight into it.
Do I need programming experience before starting LabVIEW?
No prior text-based programming is assumed. The lectures present LabVIEW as a graphical language in which wire colours and icon shapes show data types, and the first laboratory starts from no knowledge at all. You will still benefit from installing it early, following the vendor's basic training videos and practising loops and graphs on a campus computer before the hardware sessions.
What mathematics does the module use?
The notes use algebra, logarithms and simple trigonometry rather than advanced calculus. You rearrange an exponential first-order response to get a time constant, compute a magnitude ratio and phase lag from the product of time constant and angular frequency, express residuals as percentages of full scale, and convert between decimal and binary. Keeping units and percentages straight is where most errors arise.
Which topics should I revise first for the quizzes?
Start with the measurement vocabulary, because every later topic reuses it: range, sensitivity, non-linearity, offset, hysteresis, resolution and accuracy. Then practise the quantitative pieces, namely rise time to time constant, magnitude ratio, phase lag, the Nyquist-Shannon criterion and full-scale resolution.
Finish with number formats, since a wrapped signed integer is easy to recognise once you have decoded a few by hand.
How should I record data in the laboratory?
Save the raw voltages as well as any calibrated values. The lectures stress this because undoing a mistaken calibration after the event is very difficult. Keep a record of when, where, by whom, against which standard and under what conditions each calibration was made, and choose TDMS rather than CSV when a run produces large files or several runs belong together.
Questions students ask in each AERO61021 chapter
- Why run experiments when CFD and FEA are so capable?
- Simulations resolve only part of the physics and are limited by computing power, so they tend to depart from reality at the edges of a design envelope and on genuinely new problems. Experiments provide benchmark data, certification evidence and the comparison needed to validate a model, which is why most modern projects combine both. Chapter 01 · Why Engineers Experiment
- What does a rating such as 10 mV/V/N mean?
- It is a ratiometric sensitivity. For every volt of excitation the sensor produces 10 mV per newton, so at 5 V excitation it gives 50 mV per newton. Because the output depends on the supply, keep the excitation steady and record it alongside the signal so that any drift can be corrected later. Chapter 02 · Sensor Characteristics and Measurement Terms
- Why does gauge pressure change from day to day?
- A gauge sensor measures relative to the surrounding atmosphere, and atmospheric pressure varies with the weather. The same absolute pressure can therefore give different gauge readings on different days. Record the ambient pressure during each test so readings can be converted to absolute values and compared fairly. Chapter 03 · Data Types and Measurement References
- Is calibration just multiplying by one number?
- Only when the device responds linearly and the measurement is steady. Many sensors follow curved laws that need a quadratic, cubic or other fitted function, and time-varying measurements can need a full transfer function because the response changes with frequency. The safe description is a relationship between reading and quantity, established by stepping the input and recording the output across the whole range. Chapter 04 · Calibration Standards and Resolution
- Why does a slow sensor shrink an oscillating signal?
- A first-order sensor is always chasing its input. When the input reverses before the output has caught up, the output turns back early and never reaches the full peak. The faster the oscillation relative to the time constant, the smaller the fraction that survives, which is what the magnitude ratio expresses through the product of time constant and angular frequency. Chapter 05 · Sensor Response Time and Dynamic Signals
- Is sampling at exactly twice the frequency enough?
- It is the boundary for identifying frequency, not a comfortable working point. At or just above twice the signal frequency the record can still misrepresent the waveform shape, and the peak-to-peak size is unreliable. The module notes illustrate a far higher rate, around ten samples per cycle, when amplitude must be read correctly from the data. Chapter 06 · Sampling Rate and Aliasing
- How do I pick the range of a load cell?
- Estimate the largest total load first, including any rig, mounting plate or model that rests on the cell, then choose the smallest standard range that exceeds it with some margin. A much larger range keeps the device safe but spreads its full-scale uncertainty across a reading that uses only a small part of the scale. Chapter 07 · Choosing and Powering Sensors
- Why does my LabVIEW graph suddenly jump from 127 to minus 128?
- The value has been converted to a signed 8-bit integer, whose largest value is 127. One step beyond that, the leading bit becomes 1 and is interpreted with a weight of minus 128, so the trace drops to the bottom of the range and starts climbing again. Use a wider integer or a floating-point type for the quantity. Chapter 08 · Binary Number Formats in Data Acquisition
- Do I need to know a text programming language before learning LabVIEW?
- No. The lectures state that text-based programming experience is not required, because the language is graphical and built around wiring functions together. What helps instead is reading dataflow: knowing that a node runs when its inputs arrive, that inputs sit on the left of a function, and that wire colour shows the data type. Repetition does the rest, so practise on any campus computer with simulated data before the hardware sessions. Chapter 09 · LabVIEW Programming Structures
- Why does my potentiometer voltage drift when I am not touching it?
- The DAQmx channel defaults to a differential connection, which expects a positive and a negative wire. With only the signal wire connected, the negative input is left floating and the value wanders. Change the terminal configuration to RSE so the internal earth acts as the negative; the handout shows that the reading then holds steady and can reach both 0 V and 5 V. Chapter 10 · Acquiring and Saving Data with DAQmx
How to prepare for the assessments
Treat AERO61021 as one measurement chain and revise it in the order a signal travels. First, fix the vocabulary: write one line defining range, full scale, sensitivity, non-linearity, offset, hysteresis, resolution and accuracy, then attach a worked number to each from your own invented sensor.
Second, drill the short calculations until the set-up is automatic: percentage non-linearity from the largest residual, hysteresis error at the mid-range input, resolution from a percentage of full-scale output, a time constant from a quoted rise time, and the magnitude ratio and phase lag at a stated frequency.
Third, practise the judgement questions the notes keep returning to: which reference a pressure reading uses, whether a point measurement has enough spatial resolution, which load cell range covers the dead weight, and what sample rate both avoids aliasing and resolves amplitude.
Fourth, decode binary by hand, including a signed 8-bit value that has wrapped, because the notes show how plots that make no sense often trace back to a number format. Finally, rebuild the first laboratory VI from a blank block diagram on a campus computer: create the channel, set the timing, read a block of samples, plot it on a graph, apply a calibration coefficient and write the file.
If you can do that without the handout, the group project starts from firm ground.