RC Slab and Shell Assessment – MIDAS Support

Intro

Assessing a reinforced concrete slab or shell out of a plate model is a transcription job before it is an engineering one. MIDAS reports plate moments and membrane forces in each element’s own local axes, and a real mesh does not have those axes pointing the same way; the reinforcement runs in directions of its own; every quad carries five result positions and every load case another set of them; and both faces have to be assessed, by more than one theory, before the worst of it can be reported. Done by exporting tables and working them in a spreadsheet it is slow, it has to be rebuilt whenever a bar size or a load case changes, and the one number that matters ends up separated from the arithmetic that produced it. The RC Slab & Shell Assessment plugin does the whole of it inside CIVIL NX. Choose the plates, the reinforcement direction, the bars and the load cases, and it rotates every reading into the reinforcement frame, assesses it by Denton–Burgoyne, by Wood-Armer, by the Morley / Clark-Nielsen sandwich route and by the EN 1992-2 Annex LL sandwich model, and reports the governing utilisation per element — with a printable calculation report in which every derived quantity is a live Excel formula. Orthogonal and skew reinforcement are both carried, and the two sandwich routes take membrane force alongside bending, so a wall or a shell is assessed the same way a flat slab is.

Developed with

MIDAS CIVIL NX 2026 (v1.0.0)

Developed by

MIDAS IT EUROPE — manoj@midasit.com

Benefits of this plugin

Four assessment routes from one model read — Denton–Burgoyne in closed form, Wood-Armer, the Morley / Clark-Nielsen sandwich route and the EN 1992-2 Annex LL sandwich model, on both faces at every result position. The two sandwich routes carry membrane force as well as bending; Annex LL adds the cracking criterion, the concrete checks and transverse shear. Tick the routes you want; the plugin reports the one that governs and says which it was.
Slabs and shells, not only flat slabs — the Morley layer split followed by Clark-Nielsen membrane design assesses bending and in-plane force together, so a wall, a box or a curved shell is handled the same way a flat slab is. It is selectable on its own, and Annex LL sits beside it as the Eurocode form of the same idea — layer thicknesses from LL.113, the cracking criterion, the concrete strength checks and the transverse shear the inner layer has to carry.
Skew reinforcement carried through every route — one angle α between the two bar directions is set with the bars and is shared by all four methods, including the case where capacities are typed in directly rather than derived. Denton and the two sandwich routes take a skew mat as they stand. Wood-Armer produces its design moments in the actual bar directions, with its own lever arm and reinforcement force for each of them, rather than for notional bars along the reference axes. Its utilisation form stays orthogonal, which is what it is defined for — no substitute ratio is invented in its place, and Denton gives the skew capacity assessment.
Local axes are read, not assumed — MIDAS returns plate results in each element’s own axes, and node ordering means a real mesh does not have them aligned. Every element’s axis triad is worked out from its nodes and rotated into the reinforcement frame before it is assessed, so an unaligned mesh gives the same answer as an aligned one. Where the reinforcement direction has no projection into an element’s plane, the plugin names the element rather than guessing.
The two faces cannot be mixed silently — “top” follows the element normal, which follows node ordering and varies across a real model. Plates are grouped by normal direction and one orientation group is assessed at a time, instead of averaging two physically different faces into a single result.
It can align the mesh for you, and that is the only thing it writes — Preview alignment lists the beta angle every plate would take, Apply beta angles writes them so each local x runs the same way, and Restore previous puts them back. The plugin’s own results do not need it, but the MIDAS contours and tables are drawn in local axes and are not comparable across an unaligned mesh. Nothing else in the model can be changed: the ANGLE field of plate rows is the whole of the write whitelist.
Capacities derived from the bars you actually have — give the diameter, spacing, layer and cover per face and direction, with the slab depth and the material properties, and the areas, effective depths, lever arms, moment capacities and the Annex LL steel forces all follow from the EN 1992-1-1 rectangular stress block. Capacities can be typed in directly instead for a section already worked out elsewhere.
Transverse shear inside the Annex LL route — give the effective depth, the anchored longitudinal ratios in the principal shear direction, and — where links are needed — cot θ, the link fywd and the supplied Asw/s, and the clause 6.2 check runs as part of the assessment rather than as a separate hand calculation. The truss increments are shared between the faces per LL.124–127. Where the concrete fails, LL.115(c) accepts layer thicknesses you choose and applies LL.149/150 to them, keeping the bars where they are.
Every position and every case assessed, the worst reported — all five result positions per quad and every ticked load case are checked. Averaging hides peaks, so it is a tick-box rather than the default. Envelope-valued combinations are addressed with the sense they need, worked out from the whole combination tree rather than from the type of the combination alone.
Design forces as well as verdicts — the Wood-Armer design moments, the Denton load factor γ and the sandwich layer forces at every result position, as a table and as an Excel workbook with one sheet per theory that ran — Wood-Armer, Denton, Clark-Nielsen and Annex LL, for the times the numbers are what is wanted rather than a pass or a fail. Both exports say which theory produced each number.
A calculation report that can be checked — the governing location is re-run in full and written out as a printable A4 flow sheet: a numbered heading for what is being worked out, then symbol, equation, value, unit and clause on each line. Every derived cell is a live Excel formula, with the engine’s own value and a pass/fail comparison beside it in columns kept outside the print area.
A route that cannot be completed is never reported as a pass — a missing input, a section the integrated shear route does not cover — prestress, applied tension, a different National Annex — or a check that needs information a single result row does not carry is labelled INCOMPLETE in the element table and in both workbooks, with the reason beside it. The alternative, a quiet pass on a check that was never made, is the one outcome an assessment tool must not produce.
One implementation of the mathematics, checked at every start-up — the calculation library runs in the browser through Pyodide, so there is no second copy written in JavaScript that could drift from it, and the runtime version is pinned. Three checks taken from the specification run inside it each time the plugin opens; if any of them fails the plugin refuses to report numbers rather than show plausible ones.

How to use this plugin?

Install it and open the model — Apps ▸ API ▸ Plug-in ▸ My Work, and add RC Slab and Shell Assessment v1.0.0.zip. CIVIL NX must be open with the model already analysed. On the Connect tab the connection details are supplied by the host; press Connect and the plugin reports the session it is attached to. What it checked inside its own calculation core on start-up is listed beneath.
Choose the plates — on the Elements tab pick a MIDAS structure group to work within, where only part of the model is being assessed, and then an orientation group. The table underneath reports what the local axes are actually doing across that selection, which is the thing to read before deciding whether to align them.
Set the reinforcement direction — also on Elements. The angle θ is measured inside the group’s own plane, so any angle is valid by construction; a global vector can be given instead where that is how the direction is known. The diagram draws what has been set, from the values entered rather than as a generic illustration.
Align the local axes, if you want them aligned — choose global X, Y, Z or the reinforcement direction, press Preview alignment to see the beta angle every plate would take, and Apply beta angles to write them. Analyse re-runs the model afterwards, and Restore previous undoes the write. The assessment does not depend on this; the comparability of the MIDAS contours does.

[ screenshot: Elements tab — orientation group, reinforcement direction and the local-axis survey ]

Pick the load cases — on the Load cases tab filter and tick the static cases and combinations to assess. How each one is addressed is shown beside it, including whether an envelope-valued combination needs the (CB:max) sense. Choose the result positions here too: element centre only, and nodal averaging, are both off unless asked for.
Give the section and the reinforcement — on the Reinforcement tab set the slab depth, the concrete and steel grades with their partial factors, and then the bars: diameter, spacing and layer for each direction on each face, the covers, and the angle α between the two directions. Direction 1 is the reinforcement direction set on the Elements tab, and α is shared by all four methods — set it once and every route uses it. The derived capacities appear as the values are entered. Tick the methods to run — Denton, Wood-Armer, Sandwich / Clark-Nielsen and EN 1992-2 Annex LL — and leave Design forces ticked to have them computed in the same pass.

Reinforcement tab — the section, the bar arrangement and the guide diagram, which is drawn from the values entered rather than as a generic illustration.

Reinforcement tab — the capacities derived from those bars, and the four assessment routes to run.

Complete the shell inputs if a sandwich route is ticked — on the Shell inputs tab the geometry and the steel capacities are already derived from the section and the bars. Add the transverse-shear inputs where the plates carry shear: the EN recommended parameters, the effective depth, the anchored longitudinal ratios, and cot θ with the link fywd and Asw/s where links are required. LL.115(c) layer thicknesses can be given here too. What is left out is reported as incomplete rather than assumed.

Shell inputs tab — the Annex LL transverse-shear inputs, and the section quantities carried over from the reinforcement.

Assess — the Assess button sits in the footer and is live from whichever tab is open. Progress is reported as the readings are fetched and worked, and a long run can be cancelled.
Read the result, and export the working — the Results tab lists the governing utilisation per element, worst first, with the route and face that governed and the warnings that apply; anything that could not be completed says INCOMPLETE and why, rather than showing a utilisation as though the check had been made. The whole table can be copied as text. Export calculation report (Excel) writes the printable working for the governing location. On the Design forces tab, Compute design forces and Export Excel produce the design moments, the Denton γ values and the sandwich layer forces, one sheet per theory that ran.

Conclusion

The RC Slab & Shell Assessment plugin puts the slab and shell check where the model already is. There are no exported tables, no spreadsheet to rebuild whenever a bar size, a load case or a reinforcement direction changes, no assumption that the mesh’s local axes point the way the reinforcement does, and no need to drop to a hand calculation because the bars are skew or the plates carry membrane force. It reads the model, writes nothing but the beta angles it is asked for, and returns both the verdict and the working behind it — leaving the engineer’s time for the judgement rather than for the transcription.

MIDAS IT EUROPE — for clarifications, suggestions or feedback: manoj@midasit.com