Computational Engineering / Diagnosing computational models
Restraints, releases and connectivity
Three faults that look alike in the output and need completely different fixes.
Lesson review: expert verification remains open
This lesson has received a full-context editorial review and a check of its technical explanations. Selected arithmetic has been checked separately. This is not independent engineering sign-off.
Checks needed for project use
- A computational engineering reviewer must check algorithms, convergence claims, statistical assumptions and numerical examples beyond the cases explicitly recorded.
- Model Doctor lab case banks are separate from these lesson blocks and need their own full technical audit.
- Confirm mesh-quality limits and punching checks for each adopted element and design standard.
These are educational models and exercises. Check their assumptions before using the results. The full technical content and every diagram have not yet been independently verified.
Concrete and steel calculations retain their stated first-generation Eurocode teaching basis. Use the standards and amendments specified for your project; do not mix generations without an appropriate design basis.
Read the lesson register and outstanding review notesWhy this matters
Restraints, releases and connectivity all describe how parts of a model are joined — to the world, to each other, and at all. All three produce symptoms in the same family: load in the wrong place, a member carrying nothing, a bay deflecting more than its neighbours.
They need different fixes, and applying the wrong one usually makes the model run while making it less like the structure.
By the end of this lesson you should be able to
- State precisely what a restraint is and what a release is
- Recognise the symptom pattern each fault produces
- Explain why over-releasing produces a mechanism
- Explain why infinite support stiffness is usually the wrong model
- Identify coincident nodes and coincident elements from their symptoms
Restraints and releases
A support restraint relates a degree of freedom to ground or a prescribed reference and can produce a reaction. A member-end release removes a specified internal force transfer. Check which operation the software applies: mistakenly grounding beam ends can divert loads away from columns while preserving global equilibrium.
Place releases where the physical connection or hinge requires them. Subdividing a continuous member does not create new hinges. Releasing every sub-element end can introduce mechanisms; a release count is a useful warning, not proof that every such model is wrong.
Support stiffness
Horizontal restraint can change truss action and chord forces. Model the actual support detail and assess uncertain stiffness rather than assuming the force change is always the same. Rigid supports can be appropriate idealisations where their flexibility is insignificant; soil–structure interaction may instead require justified spring stiffness and sensitivity studies. An assumed 10 mm settlement is only an illustrative input, not a general foundation rule.
Connectivity
Shared nodes are one means of connection. Ties, multipoint constraints, contact and coupling definitions can also connect nonmatching meshes. Coincident geometry alone does not establish a connection. Check duplicate nodes and elements, interface constraints and the effect of any hanging nodes in the chosen formulation. Include node, element, length, volume and mass counts after copying or importing geometry.
Try it
Restraint and release laboratory
A shallow truss supported at bottom-chord level. Change how it is held and watch the load path change with it.
Supports
Deeper means the supports sit further from the neutral axis, so the arching effect grows.
Coincident node at B1
Two nodes at the same coordinate that were never merged.
- Solves?
- yes
- Equilibrium residual
- 4.71e-13 kN
- Total horizontal reaction
- 2.27e-13 kN
- Bottom chord at midspan
- 114.1 kN
as expected under vertical load
tension, as a beam
Pin and roller: statically determinate, no horizontal reaction under vertical load, and the bottom chord is in tension as a beam analogy predicts.
Try this in order
- Start with pin and roller and note the bottom chord tension. That is the force the member has to be designed for.
- Switch to pinned both ends. A horizontal reaction appears from nowhere and the bottom chord tension collapses. Nothing warns you.
- Increase the depth and watch the effect grow — the further the support sits from the neutral axis, the stronger the arching.
- Turn on the coincident node with the supports back to pin and roller, and watch the residual stop balancing.
What this shows: Over-restraint runs cleanly and changes the structure. A truss pinned at both ends below its neutral axis works as an arch, and its bottom chord loses its tension.
Predict first
A simply supported truss is modelled with pinned restraints at both ends, at bottom chord level rather than at the neutral axis. Under gravity load, what happens to the bottom chord force compared with a roller at one end?
Practice
A continuous steel beam over three spans is split into 8 elements per span, and moment releases have been applied at both ends of every element. How many moment releases are there in total?
Practice
A foundation carries 1 800 kN of unfactored load and is assumed to settle 10 mm under it. What spring stiffness should be used, in kN/m?
Check yourself
Two shell meshes touch geometrically but have no shared nodes, ties, contact or other coupling. What has been modelled?
Summary
- A restraint joins the model to the world and generates a reaction; a release does neither
- Total right and distribution wrong is a short-circuited load path
- Releases belong to members, not to the elements a member is divided into
- Over-restraint runs cleanly and changes the structure
- A sawtooth that gets worse with refinement is a line of rigid supports
- Verify shared nodes or explicit interface coupling; coincident geometry alone is insufficient