Read a simulation result beyond the color plot
Check the physical setup, native boundaries, mesh, residuals, conservation and refinement before interpreting a result.
A color plot compresses a large amount of information into an image, but it can also hide an incorrect setup. Before interpreting displacement, stress or temperature, establish what problem was solved and which evidence supports the answer. Intact associates a report with its source geometry and preserves the numerical audit context.
Start with the physical question
Specify the analyzed solid, units, material properties and boundary conditions. For structural work, identify supports and whether a load is a total force, pressure or other supported condition. For steady conduction, specify conductivity and actual thermal boundaries. A wrong boundary selection can produce a numerically tidy answer to the wrong question.
Use native face references and inspect the selected regions. If geometry changes, confirm that the study still refers to the current artifact. Do not assume a face number identifies the same surface after a rebuild.
Check the supported regime
Intact’s bounded mechanical workflows include small-strain linear elasticity, steady thermal conduction and creeping-flow Stokes. They do not establish whole-assembly contact, nonlinear mechanics, turbulence, inertial CFD or multiphysics. Whole-vehicle suspension behavior requires capabilities and inputs beyond a single-solid compression case.
Virtual Analysis is a different tool: it previews prescribed kinematics. Wheel rotation and a moving chassis do not prove traction, collision handling, suspension forces or balance.
Read the effective mesh
Record element counts, quality and actual resolution. Automatic sizing may coarsen an overly fine request to stay within computational limits. The requested element size is not necessarily the size that was solved. A mesh picture helps inspect distribution, but it does not establish convergence.
Refinement compares quantities across successive resolutions. Displacement can stabilize before local stress does. An analysis that stops at a resource budget without meeting the target remains unverified; do not relabel it converged because the final image appears smooth.
Read numerical diagnostics in context
Residuals measure how well the discrete equations were solved. Conservation checks test balances such as applied force versus reaction, or heat input versus heat output. Both are useful, but neither proves that the material values, supports or model dimensions describe the real system.
The headless result audit checks finite fields, residuals, conservation, refinement and supported regime information. Invalid numerical output is rejected. Missing refinement or out-of-regime results remain explicitly unverified or require review.
Use an independent comparison
For a uniform compression block, compare displacement with FL/(EA), stress with F/A and the summed reaction with the total applied force. The 1000 N benchmark works through this example. Agreement on that case provides bounded verification evidence; it is not a certification of arbitrary geometries.
For more complex work, select appropriate reference solutions, experimental data or independent engineering analysis. Document what remains uncertain.
Keep reporting faithful to the result
Read the deformation scale before judging shape changes. Visual amplification is not the physical displacement and does not modify the CAD. Retain units, material values, source revision, mesh information and verification status in exported results and review decks.
Report the study assumptions, effective mesh, peak quantities with units, support reaction or conservation evidence, residual, refinement status and unsupported aspects. Keep failed and unverified checks visible.
Follow the simulation tutorial before preparing your first study.
Install once.
Design through conversation.
Download Intact v0.20.0 for Windows. Install privately in Codex with your own account; CAD runs on your computer.
Follow the installation procedure →