A mechanical engineering project viva tests one thing above all else: whether you understand the engineering decisions you made — not just the software you used to make them. Examiners do not want to hear that ANSYS gave you a stress of 187 MPa. They want to know whether that stress is below the yield strength, what your safety factor is, whether your mesh was converged, and what would happen to the structure if the load increased by 20%. This guide gives you 60+ examiner questions with model answers across FEA, CFD, thermal systems, manufacturing, and machine design — built to prepare you for the technical depth a mechanical engineering viva demands.
Fig. 1 — Mechanical Engineering Project Viva: Five domain-specific question sets covering FEA, CFD, Thermal, Manufacturing and Machine Design
Mechanical engineering project viva examiners ask questions in four categories:
- Project Understanding — what engineering problem you solved, why it matters, what your specific contribution is
- Technical Depth — why you chose this material, mesh type, boundary condition, or manufacturing process over alternatives
- Results Validation — how you validated FEA/CFD results, what your safety factor is, how results compare to analytical solutions
- Fundamentals — underlying theory: stress-strain, Fourier's law, Reynolds number, manufacturing tolerances — depending on your domain
This guide covers 60+ questions with model answers across FEA Structural Analysis, CFD Fluid Dynamics, Thermal Systems, Manufacturing Processes, and Machine Design projects. Every answer follows the same structure: direct response, technical justification, honest limitation acknowledgement.
- How Mechanical Viva Examiners Think — What They Are Testing
- Universal Opening Questions — Every Mechanical Viva Starts Here
- FEA Structural Analysis Viva Questions and Answers
- CFD Fluid Dynamics Viva Questions and Answers
- Thermal Systems and Heat Transfer Viva Questions
- Manufacturing Process Viva Questions
- Machine Design and Mechanism Viva Questions
- Handling Difficult Viva Moments — Strategy Guide
- Conclusion — What Separates a Strong Mechanical Viva from a Weak One
- Frequently Asked Questions
Mechanical engineering project vivas are particularly demanding because the examiner can probe at three distinct levels simultaneously: the simulation or experimental level (what did you measure and how), the engineering judgement level (what do those results mean for the design), and the fundamental theory level (what physical principle explains that result). A student who can navigate all three levels fluently — connecting their specific ANSYS result to the von Mises yield criterion to Hooke's Law — demonstrates exactly the engineering competence that the project is designed to develop.
The question sets in this guide are built around the five most common mechanical engineering final year project domains globally: FEA structural analysis, CFD fluid dynamics, thermal and heat transfer systems, manufacturing process analysis, and machine design and mechanisms. The opening questions in Section 2 apply to all five domains — prepare those first, then prepare the domain-specific section that matches your project.
Section 01How Mechanical Viva Examiners Think — What They Are Testing
| Sr. No. | Evaluation Dimension | How They Test It | What a Strong Answer Shows |
|---|---|---|---|
| 1 | Engineering Judgement | "Is your design safe?" "What is your safety factor?" "What would happen if the load doubled?" | Student interprets results in terms of design decisions — not just reports numbers from software |
| 2 | Simulation Rigour | "Did you check mesh convergence?" "How did you validate your boundary conditions?" "What turbulence model did you use and why?" | Student applied correct simulation methodology — not just ran the software and reported outputs |
| 3 | Material Knowledge | "Why did you choose this material?" "What is its yield strength?" "How does temperature affect its properties?" | Student selected material with specific criteria and knows its key mechanical properties from memory |
| 4 | Theoretical Foundation | Follow-up questions on the physics behind the simulation — Hooke's Law, Navier-Stokes, Fourier's Law | Student understands the governing equations their simulation solves — not just the software output |
| 5 | Experimental Validation | "How did you validate your simulation?" "What is the percentage error between your model and the analytical solution?" | Student validated results against analytical solution, published data, or experimental measurement |
| 6 | Design Awareness | "What would you change in your design?" "What are the manufacturing constraints on this geometry?" | Student connects analytical results to real-world design and manufacturing implications |
Mechanical engineering examiners consistently report one failure pattern: students who report software outputs without engineering interpretation. "ANSYS shows maximum stress of 187 MPa" is a software readout, not an engineering answer. "The maximum von Mises stress of 187 MPa occurs at the fillet radius — this is below the yield strength of structural steel (250 MPa), giving a safety factor of 1.34 against first yield. The location matches the stress concentration predicted by the Kt factor for this geometry." That is an engineering answer. Every number from your simulation needs that second sentence of interpretation.
Section 02Universal Opening Questions — Every Mechanical Viva Starts Here
The first five minutes of every mechanical engineering viva follow the same structure regardless of domain. These are the most predictable questions in the entire viva — and the ones most worth preparing perfectly, because the examiner's first impression of your competence is formed entirely in this opening exchange.
Model Answer Structure (90 seconds): "My project investigates [specific engineering problem — e.g., 'the structural integrity of a CFRP composite wing spar under combined bending and torsion loading']. This is relevant because [one sentence on engineering significance — e.g., 'composite spars are replacing aluminium in aerospace and UAV applications, but their failure behaviour under combined loading is less well characterised than for isotropic metals']. I [performed FEA / conducted CFD simulation / designed and tested a prototype] of [specific component/system] using [tool/method]. My key finding was [specific result with number — e.g., 'the spar fails at the mid-span under Hashin fibre compression criterion at a load 23% lower than the Tsai-Wu prediction — indicating that the choice of failure criterion significantly affects structural safety assessment for this loading configuration']."
Model Answer: Frame around an engineering gap, not personal interest. "I chose this topic because [specific engineering problem exists — e.g., 'most published FEA studies of heat exchangers use simplified boundary conditions that do not account for maldistribution effects']. Existing analyses [name limitation]. My project addresses this by [your specific approach]. This framing — engineering gap first, your solution second — demonstrates that your topic selection was based on technical reasoning, not convenience.
Model Answer: Name three real limitations with mitigations. "First, my FEA model assumes linear elastic material behaviour — I did not model plasticity beyond the yield point. For the load range studied this is valid, but for higher loads a nonlinear material model would be needed. Second, I assumed [specific simplification — e.g., uniform temperature distribution across the cross-section] — in reality [the actual behaviour]. Third, I validated only against published data for a similar geometry — experimental validation on the actual component would strengthen the results. These are limitations I acknowledged in my report and would address with more time."
Model Answer: One honest, specific change. "I would perform a more thorough mesh convergence study earlier in the project — I ran convergence at two mesh densities before settling on my final mesh, but a more systematic study across four or five densities would give me higher confidence in the result. I would also [second specific change — e.g., 'include a fatigue analysis in addition to the static strength analysis, since the component experiences cyclic loading in service']. I learned the importance of this from my literature review but did not have enough time to implement it within the project timeline."
Section 03FEA Structural Analysis Viva Questions and Answers
FEA project vivas are the most common mechanical engineering viva type — and the most likely to expose gaps between what the software reported and what the student actually understands. Examiners specifically probe mesh quality, boundary condition justification, failure criterion selection, and result interpretation. Know your element type, your mesh convergence data, your maximum stress location, and your safety factor — all from memory, without opening your report.
Model Answer: "I used [element type — e.g., SOLID187 tetrahedral elements / SHELL181 shell elements / BEAM188 beam elements] because [specific reason — e.g., 'SOLID187 10-node tetrahedral elements are well-suited to the complex curved geometry of my component — they conform to curved surfaces better than hexahedral elements without requiring a structured mesh, and the midside nodes provide quadratic displacement interpolation that captures bending stress gradients accurately']. I considered [alternative — e.g., hex elements] but [specific reason for not choosing it — e.g., 'the irregular geometry would require significant manual partitioning to generate a structured hex mesh, adding preparation time without a proportional accuracy benefit for this geometry']."
Model Answer: "Mesh convergence is the process of progressively refining the finite element mesh until the solution stops changing significantly — confirming that the result is independent of mesh density, not an artefact of mesh coarseness. I performed convergence by running four simulations at mesh sizes of [e.g., 8mm, 4mm, 2mm, and 1mm global element size]. The maximum von Mises stress changed by [e.g., 18%, 6%, 1.8%] between successive refinements. I selected the [e.g., 2mm] mesh as my final model because the change between 2mm and 1mm was below 2% — this represents a converged solution. The converged stress value was [X MPa] with [Y] elements total."
Model Answer: "Von Mises stress (distortion energy criterion) is the appropriate failure criterion for ductile materials — it predicts yielding when the deviatoric strain energy reaches the value that causes yielding in a uniaxial tension test. My material is [structural steel / aluminium alloy] — a ductile metal — so the distortion energy theory applies. The maximum von Mises stress in my model is [X] MPa, compared to the material yield strength of [Y] MPa, giving a safety factor of [Y/X]. If I had been analysing a brittle material [e.g., cast iron or ceramic], I would have used the maximum principal stress criterion instead, since brittle materials fail by fracture rather than yielding."
Model Answer: "I validated my FEA model using [validation method]. For a simply supported beam under central point load, the analytical deflection is δ = PL³/48EI. My FEA result for the same geometry and loading gave δ = [FEA value] mm versus the analytical solution of [analytical value] mm — a difference of [%]. This [X%] agreement confirms that my mesh, element type, and boundary conditions are correctly set up. I then applied this validated setup to the actual geometry — where no closed-form analytical solution exists — with confidence that the model is producing physically correct results."
Model Answer: "I applied [boundary conditions — e.g., 'a fixed support at both flange ends — all six degrees of freedom constrained — representing the bolted connection to the main structure; a distributed pressure load of [X] MPa on the top face representing the design load']. I justified the fixed support condition because [specific reason — e.g., 'the bolted flange connection provides sufficient rotational stiffness that a pinned approximation would underestimate the actual constraint']. I performed a sensitivity check by running the simulation with both fixed and pinned supports — the maximum stress differed by [X%], confirming that the boundary condition choice has [significant / minor] influence on the result."
Model Answer: "A stress concentration factor (Kt) is the ratio of the maximum stress at a geometric discontinuity to the nominal stress away from the discontinuity — it quantifies how much a feature like a hole, fillet, notch, or step amplifies local stress. In my model, the highest stress concentration occurs at [specific location — e.g., the fillet radius between the flange and the web] where the stress contour shows a peak of [X] MPa compared to the nominal stress of [Y] MPa — giving a stress concentration factor of approximately [X/Y]. I used a fillet radius of [value] mm to limit Kt — a smaller radius would increase the concentration, potentially causing fatigue failure under cyclic loading."
Model Answer: "Linear static FEA assumes: (1) linear elastic material behaviour — stress is proportional to strain via Hooke's Law; (2) small displacements — deformations are small enough that the geometry does not change significantly; (3) static loading — no inertia or damping effects. My project used linear static FEA because [justification — e.g., 'the maximum stress is below the yield strength at all load cases, so linear elastic behaviour is valid; and the maximum deflection is less than 5% of the characteristic dimension, satisfying the small displacement assumption']. Nonlinear FEA is required when [any of these three assumptions is violated — e.g., when loads cause plastic deformation, when large deflections change the load path, or when contact conditions change during loading]."
Section 04CFD Fluid Dynamics Viva Questions and Answers
CFD vivas probe understanding of the governing equations, turbulence model selection, boundary condition setup, and most critically — validation. Examiners are acutely aware that CFD results can look plausible even when the setup is incorrect — which is why validation against experimental or analytical data is non-negotiable, and why they ask about it specifically.
Model Answer: "I used the [k-ω SST / k-ε realizable / Spalart-Allmaras] turbulence model because [specific reason tied to your flow type]. k-ω SST (Shear Stress Transport) was selected for my [external aerodynamics / aerofoil / wind turbine] project because it combines the k-ω model's accuracy in the near-wall region (where adverse pressure gradients and boundary layer separation are important) with the k-ε model's accuracy in the free stream — making it the most widely validated model for flows with separation, which my project involves. [Alternative, e.g., k-ε] was rejected because it over-predicts turbulent viscosity in adverse pressure gradient regions, leading to delayed or absent separation prediction — which would give non-physical lift and drag results for my geometry."
Model Answer: "The Reynolds number Re = ρVL/μ for my flow condition is [calculated value — show calculation: e.g., Re = (1.225 kg/m³ × 15 m/s × 0.3 m) / (1.789×10⁻⁵ Pa·s) = 307,000]. This is [above / below] the critical Reynolds number for [flat plate transition ~5×10⁵ / pipe flow ~2,300], indicating that the flow is [turbulent / laminar / transitional]. This determines my choice of turbulence model — a turbulent flow requires a turbulence model; a laminar flow (Re below ~2,300 for pipe, below ~5×10⁵ for external flow) can be solved with the laminar Navier-Stokes equations without a turbulence closure model."
Model Answer: "I validated my CFD model against [published experimental data / analytical solution / published numerical benchmark]. For my NACA 4412 aerofoil at 8° angle of attack, I compared my ANSYS Fluent Cl result of [X] against the NACA Technical Report 824 wind tunnel data of [Y] — a difference of [Z%]. This [Z%] agreement is within the acceptable range for k-ω SST at this Reynolds number and angle of attack, based on published benchmark studies. I then used this validated setup to study [your actual investigation — geometry variation, modified condition] with confidence that the solver settings are producing physically correct results. Without this validation step, my results would have no credibility regardless of how clean they look."
Model Answer: "At the inlet: [velocity inlet / pressure inlet / mass flow inlet] with [value and units — e.g., velocity of 15 m/s, turbulence intensity of 1%, turbulent length scale of 0.01 m based on 1% of hydraulic diameter]. I chose [velocity inlet / pressure inlet] because [specific reason — e.g., 'the inlet velocity is known from the operating condition; pressure inlet is more appropriate when total pressure is known but velocity distribution is not']. At the outlet: [pressure outlet] with gauge pressure of 0 Pa — representing atmospheric pressure at the outlet plane, which is standard for external aerodynamics and internal duct flows where the outlet is far enough from the region of interest that backflow is not expected. The domain extended [X chord lengths / hydraulic diameters] upstream and [Y] downstream to minimise boundary condition influence on the region of interest."
Model Answer: "y+ is a dimensionless wall distance that characterises how well the near-wall mesh resolves the turbulent boundary layer — specifically, it measures the first cell centroid distance in viscous wall units. For wall-resolved simulations using k-ω SST without wall functions, the target is y+ ≈ 1 — the first cell must be within the viscous sublayer (y+ < 5). For simulations using wall functions with the k-ε model, y+ should be between 30 and 300 — within the log-law region. In my simulation, I targeted y+ ≈ 1 because I used k-ω SST with low-Reynolds corrections to capture the boundary layer accurately near the suction surface where separation is expected. My actual y+ contour plot shows a maximum of [value] on the suction surface — [within / slightly above] the target, which I discussed as a limitation."
Section 05Thermal Systems and Heat Transfer Viva Questions
Thermal project vivas test understanding of the three heat transfer modes — conduction, convection, and radiation — and whether the student can identify which mode dominates in their system and why. Examiners expect students to know the governing law for each mode (Fourier's, Newton's, Stefan-Boltzmann) and to connect their simulation results to those physical laws.
Model Answer: "The three modes are: (1) Conduction — heat transfer through a solid or stationary fluid via molecular interaction; governed by Fourier's Law: q = −kA(dT/dx). (2) Convection — heat transfer between a surface and a moving fluid; governed by Newton's Law of Cooling: q = hA(Ts − T∞). (3) Radiation — heat transfer by electromagnetic emission; governed by Stefan-Boltzmann Law: q = εσA(Ts⁴ − T∞⁴). In my [heat exchanger / heat sink / furnace] project, [mode] dominates because [specific physical reason — e.g., 'forced convection dominates because the high fluid velocity (Re = [value]) produces a convective heat transfer coefficient h = [value] W/m²K — approximately [X] times larger than the equivalent conduction resistance through the wall thickness']."
Model Answer: "The overall heat transfer coefficient U combines all thermal resistances between the two fluids in a heat exchanger — hot-side convection resistance (1/h_hot), wall conduction resistance (t/k_wall), and cold-side convection resistance (1/h_cold) — in series: 1/U = 1/h_hot + t/k_wall + 1/h_cold. In my shell-and-tube heat exchanger, I calculated: h_hot = [value] W/m²K (shell-side, using Kern method), wall conduction resistance = t/k = [thickness/conductivity] = [value] m²K/W, h_cold = [value] W/m²K (tube-side, Dittus-Boelter equation for turbulent flow, Nu = 0.023 Re⁰·⁸ Pr⁰·⁴). Overall: U = [calculated value] W/m²K. I validated this against my CFD-computed heat transfer rate — the agreement was within [X%]."
Model Answer: "The effectiveness-NTU (Number of Transfer Units) method is used for heat exchanger analysis when the outlet temperatures are unknown — which makes the LMTD (log mean temperature difference) method unsuitable because LMTD requires both inlet and outlet temperatures of both streams. Effectiveness ε = actual heat transfer / maximum possible heat transfer; NTU = UA/C_min where C_min is the smaller of the two fluid heat capacity rates. I used the ε-NTU method for [specific reason — e.g., 'my design problem specified inlet temperatures and required me to find outlet temperatures and required heat transfer area — exactly the scenario where ε-NTU is the appropriate method']."
Model Answer: "Thermal resistance R = ΔT/Q is the temperature difference per unit heat flow — analogous to electrical resistance (voltage/current). It allows heat transfer circuits to be analysed using circuit analogy: resistances in series add directly (R_total = R₁ + R₂ + ...), resistances in parallel add as reciprocals. In my [heat sink / composite wall / multi-layer insulation] project, I analysed the thermal circuit with [X] resistances in series: [list them — e.g., contact resistance at base, conduction through aluminium fin, convection from fin surface to air]. The dominant resistance is [which one] at [value] K/W — [percentage] of total — meaning [engineering implication — e.g., 'improving contact conductance at the base would give the largest reduction in junction temperature']."
Section 06Manufacturing Process Viva Questions
Model Answer: "I selected [process — e.g., CNC milling / injection moulding / 3D printing / casting] because [specific criteria — e.g., 'CNC milling was selected for my aluminium bracket because the required dimensional tolerance of ±0.05 mm and surface finish of Ra 1.6 µm cannot be achieved by casting or powder metallurgy without post-machining; CNC milling achieves both in one operation']. I evaluated [alternative processes] but rejected them because [specific limitations — e.g., 'casting would require a pattern and post-machining to meet tolerance; 3D printing in the available material could not achieve the required surface finish without post-processing that would exceed the project timeline']."
Model Answer: "Surface roughness Ra is the arithmetic mean deviation of the surface profile from the mean line — it quantifies how smooth or rough a machined surface is. Ra is measured in micrometres (µm). The Ra value I achieved was [value] µm using [process and cutting parameters — e.g., 'end milling at 2000 rpm, 100 mm/min feed, 0.5 mm depth of cut with a 6mm carbide end mill']. This meets the [drawing requirement / functional requirement] of Ra ≤ [spec] µm because [functional reason — e.g., 'the mating surface requires Ra ≤ 3.2 µm to achieve adequate sealing with the gasket']. I measured surface roughness using [profilometer model] with a sampling length of [value] mm."
Model Answer: "GD&T (Geometric Dimensioning and Tolerancing) is a system for defining and communicating engineering tolerances that specifies not just size but also geometric form, orientation, location, and runout of features — using standardised symbols on engineering drawings. In my project, I applied GD&T symbols for [specific features — e.g., 'flatness tolerance of 0.05 mm on the mating face to ensure proper sealing; cylindricity tolerance of 0.02 mm on the bearing bore to ensure correct bearing fit; position tolerance of ±0.1 mm for the bolt hole pattern relative to the datum axis']. Without GD&T, the drawing would specify only size tolerances — which does not fully constrain the geometry for functional assembly."
Model Answer: "Tolerance is the total permissible variation in a dimension — the difference between maximum and minimum allowable size. Fit describes the relationship between a shaft and a hole when assembled — determined by the relative sizes of both. The three fit types are: clearance fit (shaft always smaller than hole — for rotating or sliding parts), interference fit (shaft always larger than hole — for force-fitted or press-fitted assemblies requiring no relative motion), and transition fit (can be either clearance or interference depending on actual sizes — for precision assemblies requiring accurate location). In my [gearbox / bearing housing / coupling] project I used [specific fit — e.g., H7/k6 transition fit for the bearing inner race — providing accurate location with minimal clearance while allowing assembly without press-fitting equipment']."
Section 07Machine Design and Mechanism Viva Questions
Model Answer: "Material selection involved four criteria specific to my application: (1) Mechanical requirement — [e.g., minimum yield strength of 250 MPa to achieve the required safety factor of 2.0 under maximum design load]; (2) Environmental requirement — [e.g., corrosion resistance for the operating environment]; (3) Manufacturing requirement — [e.g., weldability / machinability for the chosen fabrication method]; (4) Cost and availability. I evaluated [list candidate materials] using a weighted criteria matrix. [Selected material] scored highest because [specific technical reason — e.g., 'AISI 304 stainless steel provides the required strength, excellent corrosion resistance in the salt-water environment, and good weldability — at a cost premium over mild steel that is justified by the extended service life in the corrosive environment']."
Model Answer: "The factor of safety (FoS) is the ratio of the material's failure strength to the maximum stress in the component under design loading — it quantifies the margin between the operating condition and failure. FoS = Yield Strength / Maximum Stress (for ductile materials under static loading). I used a factor of safety of [value, e.g., 2.5] because [specific justification — e.g., 'the loading is not precisely known (uncertainty in dynamic load factor), the consequences of failure are significant, and the material properties have some variability — a FoS of 2.5 is standard for structural steel components under variable loading per [relevant design standard]']. My calculated maximum stress of [X] MPa against a yield strength of [Y] MPa gives an actual FoS of [Y/X] — [above / at / slightly below] the design target."
Model Answer: "Fatigue failure occurs when a component fails under cyclic loading at stress levels below the static yield strength — due to progressive crack initiation and propagation at stress concentration sites. I assessed fatigue life using [method — e.g., S-N curve approach (stress-life method) / strain-life method / fracture mechanics]. Using the S-N curve for [material], the endurance limit Se = [value] MPa (the stress amplitude below which fatigue life is theoretically infinite). My design stress amplitude of [X] MPa is [below / above] Se, indicating [infinite life / finite life of approximately N cycles estimated from the S-N curve]. I applied Miner's Rule for the variable amplitude loading spectrum, with a cumulative damage ratio of [D value] — [below 1.0 confirms the design is safe against fatigue / above 1.0 indicates redesign is needed]."
Section 08Handling Difficult Viva Moments — Strategy Guide
| Sr. No. | Difficult Situation | Wrong Response | Correct Response |
|---|---|---|---|
| 1 | Cannot remember a specific material property value | Guessing a number without acknowledging uncertainty | "I do not have the exact value from memory — I know it is in the range of [approximate range] for this material class. In my report I used [correct value from your datasheet] — I can refer to Table [X] if needed." Do not invent numbers. |
| 2 | Examiner questions your safety factor choice | "I just used the default value" or becoming defensive | "I selected FoS = [value] based on [specific reasoning: loading uncertainty, consequence of failure, material variability, design standard reference]. If [examiner's concern] were present, I would increase the FoS to [value] to account for it." |
| 3 | Did not perform mesh convergence | Pretending you did, or claiming it was unnecessary | "I acknowledge that a formal mesh convergence study was not completed — this is a limitation of my project. I selected my mesh density based on [element size guidelines from literature / software recommendations], but I cannot claim convergence without the systematic study. Future work would include this." |
| 4 | Examiner asks about a theory you have forgotten | Attempting to recall and producing an incorrect statement | "I recall that [concept] relates to [broad area] but I am not confident enough in the specific formulation to state it accurately right now. What I can say is that in my project, the relevant equation I used was [equation you did use], which comes from [theory area]." |
| 5 | Your experimental or simulation result seems wrong | Immediately agreeing it is wrong under pressure | "My result was [value], obtained using [validated methodology]. I am confident in the methodology. If there is a specific physical reason why you would expect a different result, I would be interested to understand — it may relate to [specific assumption in my model]." |
| 6 | Examiner asks about manufacturing aspects of a simulation-only project | "My project was simulation only, so I did not consider manufacturing" | "My project focused on computational analysis, but I considered manufacturing implications: [specific examples — e.g., 'the minimum fillet radius in my optimised geometry is 2mm, which is achievable with standard end milling; the material I selected has good machinability; the optimised topology would require support structures if 3D printed']." |
Section 09Conclusion — What Separates a Strong Mechanical Viva from a Weak One
The single clearest indicator of a strong mechanical engineering viva is the ability to move fluently between three levels of discussion: the specific result ("my maximum von Mises stress is 187 MPa"), the engineering interpretation ("this gives a safety factor of 1.34 against first yield, located at the fillet radius as predicted by the stress concentration factor for this geometry"), and the fundamental principle ("this is consistent with the distortion energy theory — von Mises stress integrates all stress components into a single equivalent value for comparison against uniaxial yield data"). Students who can operate at all three levels in a single answer consistently receive the highest viva marks, regardless of project complexity.
Three preparation practices produce strong mechanical vivas reliably. First, memorise your key numbers — maximum stress, yield strength, safety factor, mesh element count, convergence percentage, and maximum displacement or temperature. These are asked in almost every FEA, CFD, or thermal viva. Second, for every major decision in your report — material choice, element type, turbulence model, boundary condition, manufacturing process — prepare a three-part answer: what alternatives you considered, what selection criterion you applied, and how you validated the choice. Third, revise the fundamental governing equation for your domain — Hooke's Law, Navier-Stokes (conceptually), Fourier's Law, Newton's Law of Cooling — because examiners frequently ask "what equation is your simulation actually solving?"
Most importantly: your safety factor is the most important number in your entire project report. An examiner in a mechanical engineering viva will always ask whether your design is safe — not whether your software converged. "Yes — my safety factor is 2.3 against yield, with the critical location at the fillet where I applied a localised mesh refinement" is the answer that earns full marks. "ANSYS shows 187 MPa" is the answer that earns the next question: "And what does that mean?"
Can explain entire project in 90 seconds without notes ✓ — Know maximum stress/deflection/temperature from memory ✓ — Know material yield strength and UTS from memory ✓ — Know safety factor and can calculate it instantly ✓ — Can explain mesh convergence methodology step by step ✓ — Can justify every boundary condition with specific physical reasoning ✓ — Know governing equation for your domain (Hooke's, NS, Fourier's) ✓ — Have validation data ready (% error vs analytical/experimental) ✓ — Prepared 3 honest limitations with specific mitigations ✓ — Know page numbers of mesh convergence table, stress contour, results table ✓
Section 10Frequently Asked Questions
Four categories: project understanding (why this topic, what engineering problem it solves), technical depth (why this material/mesh/boundary condition over alternatives), results validation (how you validated FEA/CFD results, what your safety factor is), and fundamentals (Hooke's Law, Reynolds number, Fourier's Law depending on your domain).
Three-part answer: alternatives you considered (Abaqus, SolidWorks Simulation), specific selection criterion for your project (material library, composite modelling capability, solver type), and validation result (% agreement between your model and the analytical solution for a simple geometry). Never say "it is the most popular software."
Progressively refining mesh until the solution stops changing significantly. State your four mesh densities, the percentage change in maximum stress between each refinement, and the mesh size where change dropped below 2% — that is your converged mesh. State the element count of your final model.
A scalar equivalent stress combining all components, derived from distortion energy theory — a ductile material yields when von Mises stress equals its uniaxial yield strength. State your maximum value, compare it to yield strength, and give your safety factor. Know the critical location from memory.
Static analysis assumes loads applied slowly enough that inertia and damping are negligible — equilibrium is reached at each load state. Dynamic analysis accounts for inertia (mass × acceleration) and damping — required for impact loads, vibration, or when loading frequency approaches natural frequency. Justify which applies to your project with specific loading characteristics.
Three acceptable responses: partial answer using related knowledge, honest acknowledgement with reasoning attempt, or limitation acknowledgement with a proposed future approach. Never guess numerical values — examiners know when a student is fabricating, and it destroys credibility for the rest of the viva.
Revise fundamentals for your domain: FEA — Hooke's Law, von Mises criterion, boundary condition types; CFD — Navier-Stokes (conceptual), Reynolds number, turbulence models; Thermal — Fourier's Law, Newton's cooling, Stefan-Boltzmann, effectiveness-NTU; Manufacturing — tolerances, surface finish Ra, GD&T basics. Every examiner question connects to one of these foundations.
Printed bound report with sticky tabs on mesh convergence table, boundary conditions diagram, stress contour plot, and results table. Printed material datasheet. Physical prototype if your project has one. Know your safety factor, maximum stress, and mesh element count from memory — these are asked directly in almost every mechanical viva.
Viva questions, model answers, and defence strategy in this guide reflect current mechanical engineering project assessment practices at universities globally. Question sets and answer frameworks are based on examiner evaluation patterns across FEA structural analysis, CFD fluid dynamics, thermal systems, manufacturing processes, and machine design project domains. Updated June 2026.
- The Complete Guide to Engineering Project Viva 2026
- 50 Most Common Engineering Project Viva Questions and Answers 2026
- CS Engineering Project Viva Questions and Answers 2026
- How to Introduce Your Engineering Project in the First 60 Seconds
- Mechanical Engineering Final Year Project Ideas 2026
- How to Write a Methodology Chapter for Engineering Projects 2026
- Engineering Project Report Format Guide 2026
- 200+ Final Year Engineering Project Ideas 2026 — All 18 Branches