For Rhino & Grasshopper

Structural analysis for Grasshopper

Stereom is simulation software combined with implicit modelling. It analyses implicit geometry directly, without heavy meshing. A free, GPU-accelerated solver.

Get notified See the roadmap

Not yet available to download. Sign up and we will email you at release.

A long, thin part cut lengthwise, its TPMS interior coloured by von Mises stress from a Stereom solve

The von Mises stress legend for the banner: 1.14 MPa minimum to 205 MPa maximum, against a 230 MPa limit

The workflow

Implicit, end to end

Start from a Grasshopper template, or build your own from implicit generation. Breps and meshes from other CAD convert in, and edit with implicit tools.

Mesh only at the end, once a solution is found.

The Grasshopper canvas as a Stereom definition is built: Domain, Restraint, Force and Material components wired into Solve

The Rhino viewport through one solve: the lattice part, then its load and restraint vectors, then the solved stress field, then a section through it

Setup

Common inputs

Four components define the problem. Material presets ship pre-loaded, including alloys used in additive manufacturing.

Domain: The geometry to solve on.
Restraint: The regions held in place.
Force: The load, and where it lands.
Material: A preset, or your own stiffness and density.

A schematic walking the four inputs a Stereom analysis needs — Domain, Restraint, Force and Material — then all four wired into the Solve component

Grading

Field-graded lattices

Use Isopod’s implicit toolset to design optimised structures, then simulate directly on the fields. FieldView shows the grading as you work.

Modify the implicit and the display keeps up: GPU raymarching, no rebuild between edits.

A graded TPMS beam driven by a field: the grading curve is edited, the field vectors update, and the lattice follows

Benefits of implicit

No meshing

Solving straight from an implicit model skips the step that costs the most and breaks the most often. Iteration gets shorter.

The mesh-free solver runs locally. No cloud, no credits, no queue. Design and iterate on your own hardware without going over budget.

A connecting rod split down its axis: the left half meshed with conforming elements that stall and fail quality checks at a thin web, the right half covered by an implicit background grid that refines through five passes and solves, then lifting into a three-dimensional view with the stress field carried through the thickness of the part — One part, split down its axis. Left: what meshing costs. Right: the same geometry, solved as-is.

Convergence

Checking the answer

Converge refines the resolution until the outputs settle inside the tolerance you set.

You get the remaining error, not just an answer. A wall-clock budget and a resolution ceiling stop it.

A convergence run drawn as a chart: the error estimate falling with each refinement until it crosses the tolerance and stops

−4.0% → ~0.2%: vs CalculiX 2.22 on a gyroid lattice: raw at 182k DOF, then extrapolated
~1e-10: Patch test: a linear field recovered to machine precision

Where it runs

On your machine

The solver runs on your own GPU, with an automatic CPU fallback. Nothing is uploaded and nothing is metered.

Your geometry stays on your workstation, and the only thing between an idea and an answer is how fast you can change the model.

A curved bracket with a graded lattice core, coloured by displacement magnitude from 0 to 2.14 mm

Roadmap

Coming soon

The first release covers static structural analysis, and the functionality built around that type of study.

Future releases target more study types, optimization, computational geometry and automated workflows. Stereom is under active development.

In this build

Linear static stress and displacement
Implicit solids and lattices
Isotropic materials, with presets
Convergence solver with error reporting
GPU (CUDA), with automatic CPU fallback
Windows, in Rhino 8

Coming

Multiple load cases
Gravity and body loads
Nonlinear analysis
Modal and thermal analysis
Topology optimization
Field-graded optimization

Not planned: fatigue and as-built defects, the manufacturing effects a nominal-geometry solve cannot see. The full roadmap.

The name

Bio-inspired

Efficient structural lattices are found throughout nature. With advances in additive manufacturing, we can finally mimic them in engineering.

Stereom (STERR-ee-um) is the graded calcite lattice in the skeletons of sea urchins and starfish. The same geometry turns up in a fossil dated to 385 million years. The whole story.

A block of sea urchin stereom imaged in grey, joined to a computer-simulated Diamond surface in yellow; the two lattices continue into each other almost seamlessly — Jessop et al. (2024), J. R. Soc. Interface 21(212): 20230597. CC BY 4.0. Cropped.

Be first to solve

Core features will be free to use. Sign up now and your free account is made with the same click.

1InstallFrom the Rhino package manager, or by hand.
2ActivateA one-time free account, the first time you solve.
3OptimizeIterate on the model, and solve again.

Get notified at release