# Stress Linearization query **URL:** \ **Category:** Feature Requests\ **Created:** [September 24, 2026, 11:28pm UTC](https://prepomax.discourse.group/t/stress-linearization-query/3697 "2026-09-24T23:28:51Z")\ **Posts on this page:** 4\ **Page:** 2
**Author:** ![FEAnalyst](https://yyz2.discourse-cdn.com/free1/user_avatar/prepomax.discourse.group/feanalyst/32/688_2.png) [@FEAnalyst](https://prepomax.discourse.group/u/FEAnalyst)\ **Post date:** [October 1, 2026, 8:31am UTC](https://prepomax.discourse.group/t/stress-linearization-query/3697/22 "2026-10-01T08:31:19Z")
In fact, I would really appreciate it if someone submitted an AI-assisted PR to fix the issue with stress linearization in FreeCAD, but it’s not a small task: [FEM: Stress linearization is wrong (Requires retooling the code) · Issue #20522 · FreeCAD/FreeCAD · GitHub](https://github.com/FreeCAD/FreeCAD/issues/20522) ---
**Author:** ![aayestaran](https://yyz2.discourse-cdn.com/free1/user_avatar/prepomax.discourse.group/aayestaran/32/272_2.png) [@aayestaran](https://prepomax.discourse.group/u/aayestaran)\ **Post date:** [October 1, 2026, 12:03pm UTC](https://prepomax.discourse.group/t/stress-linearization-query/3697/23 "2026-10-01T12:03:44Z")
Hi Sergio!, Like you, I have never done an elastic DBA using shells. In my opinion they are useful for a quick code stress assessment away from any discontinuity, but the critical zones are usually near a discontinuity, and there the result depends greatly on how the junction is modelled to represent the right stiffness of that zone. In the past I have reviewed some reports using this approach according to ASME, mostly on the fitness-for-service side and on storage tanks. In the process piping field the stress linearization in shells is more common, because specialized commercial software like Paulin Research Group’s tools uses a template system to build the model and postprocess according to ASME with the shell stress linearization approach directly. Even in those tools, the limitations come from shell theory itself, not from the software, so we are not exempt from them whether we model is in CalculiX, FEMaster or Abaqus, ansys… About reference reports, I can’t share the ones I reviewed because they are confidential, and honestly they wouldn’t help much: they were very simple, mostly tables and figures with the calculated magnitudes, with nothing useful from a modelling point of view, following standard and recommended practices in FEA modelling. The public reference I would look at is WRC Bulletin 429 by Hechmer and Hollinger, which is the background for Annex 5-A and discusses where shell and solid results diverge, but as I said, shells in this field are not my first choice in elastic DBA. Regards, Asdrúbal (btw, I’m also testing FEMaster for its true shell formulation and arc-length method, and so far it hasn’t disappointed me to the moment) ---
**Author:** ![FEAnalyst](https://yyz2.discourse-cdn.com/free1/user_avatar/prepomax.discourse.group/feanalyst/32/688_2.png) [@FEAnalyst](https://prepomax.discourse.group/u/FEAnalyst)\ **Post date:** [October 1, 2026, 12:51pm UTC](https://prepomax.discourse.group/t/stress-linearization-query/3697/24 "2026-10-01T12:51:36Z")
If you use shells, stress linearization can be skipped, e.g., for DBA if you just change the section point from which stress output is taken in Abaqus. Unfortunately, that’s not possible with CalculiX due to its fake shells (apart from using OUTPUT=2D to get purely membrane stresses vs keeping OUTPUT=3D to get membrane + bending stresses). Only the user shell element US3 supports stresses at different locations (SPOS, SMID, SNEG). Stress linearization in Abaqus is mostly used in 2D solid axisymmetric models, but is also applicable to 3D continuum solid elements. Abaqus doesn’t support stress linearization for shells (even continuum ones). But through-thickness result variation plots can be created. ---
**Author:** ![TKemeny86](https://avatars.discourse-cdn.com/v4/letter/t/58f4c7/32.png) [@TKemeny86](https://prepomax.discourse.group/u/TKemeny86)\ **Post date:** [October 1, 2026, 2:08pm UTC](https://prepomax.discourse.group/t/stress-linearization-query/3697/25 "2026-10-01T14:08:06Z")
In DBA, SCL is just one part of the story. To be honest, I used it before a lot but diving more deeper into fitness for service or special equipment, the SCL generates more questions as it answers. Roughly 8 months before, I have a large dispute with the authorities about a reactor, subjected to creep loads and suffered from massive corrosion. Long story short, we came to the conclusion, we - local authorities in Germany and I - would follow EN 13445-3, Annex B in case by case basis. Having the 5% elongation limit is more straightforward then arguing about what I primary or secondary stress… I’ve been discussing some folks from ASME Section VIII, Div 2, the new Div 2 - 2027 would go more in the FEA direction. The recent changes - integrating more stuff from Div 1 to Div 2 and remove level 1 and 2 categories go also is direction… [Previous page](https://prepomax.discourse.group/t/stress-linearization-query/3697.md?page=1)