When meshing in Mechanical, there are Linear and Quadratic options in the meshing settings. If translated literally, they can be called linear elements and quadratic elements; conceptually, they can be understood as low-order elements and high-order elements.
Linear elements or low-order elements mainly use linear functions as shape functions. Displacement in the mesh region between nodes varies linearly with distance between nodes. Linear elements cannot capture bending.
Second-order elements or high-order elements mainly use nonlinear functions as shape functions. Displacement between nodes is interpolated using high-order polynomials. These elements have mid-side nodes, and element edges are typically composed of three nodes rather than two nodes. Second-order elements are more suitable for representing complex geometries and bending deformation. However, correspondingly, due to the increase in nodes, the number of equations to be solved increases significantly, and computational resource consumption increases.
In practical use, if they are solid elements, low-order tetrahedrons make structural models too stiff and should generally be avoided.
The figure below shows the comparison relationship between computational balance and element shape and order.
When users specify the corresponding order, if they want to view specific element types, they can check in the solution information in Mechanical. The figure below shows the comparison of shapes between common solid element low-order element 185 (8 nodes) and high-order element 186 (20 nodes).
The figure below shows the comparison of result accuracy between low-order elements and high-order elements:
The summary comparison table of differences between low-order and high-order elements is as follows:
Low-order Elements vs High-order Elements:
– Linear shape functions vs Nonlinear shape functions
– Stress state mostly constant within a single element vs Stress varies linearly within a single element
– Cannot accurately represent curved edges and surfaces vs Can accurately represent curved edges and surfaces
– Highly sensitive to element distortion vs Relatively insensitive to element distortion
– Usually only acceptable when only nominal stress results are of concern vs Recommended if accurate stress distribution is of concern
– Requires large number of elements to resolve high stress gradients vs Usually gives better results than low-order elements, often using fewer elements
– Usually lower computational cost (faster simulation runs) vs Usually higher computational cost
Shuige suggests the following strategy for practical operations:
1. If it’s just for quickly evaluating trends in response and not particularly concerned with specific values, low-order elements can be used to obtain faster calculation speed.
2. For solid elements, if tetrahedrons are unavoidable when meshing, high-order elements should be used to avoid an overly stiff model.
3. High-order elements can be used in areas of interest in the model, and low-order elements in non-critical areas—combining high and low orders with targeted application.
4. For the first calculation, low-order elements can be used entirely to obtain overall trends. If trends and response values are within acceptable ranges, switch to high-order elements to obtain more precise results.
5. If computational resources are sufficient and time is not urgent, go with high-order elements.
For simulation business cooperation, please click: Business Services
For more original courses by Shuige, please click: Shuige’s Original Course List – Up to September 2025
For more finite element learning resources, please visit Shuige’s blog: FSCAE.COM
Best regards,
ANSYS Structural Institute
November 17, 2025
No comments yet