Case study · Aerospace & motorsport · Aether generative design

Velocity Motorsport

Cut a suspension upright's mass by a third — qualified, on the first print.

Headline

First print

validated and producible, no DfM rework

In-the-loop

FEA on every candidate

Pareto

mass vs. stiffness, not one answer

0

post-DfM redesign loops

Background

Velocity runs an aggressive parts-development cycle where every gram and every race weekend counts. Their generative-design workflow was a CAD add-on that proposed organic shapes a separate FEA seat then had to validate — and the two rarely agreed, so most generated parts came back for a manual redesign.

The challenge

The generator optimized geometry against a simplified load case, then the real FEA found stress concentrations the generator never saw. Add the additive-manufacturing constraints — overhang, supports, minimum wall — and a 'finished' generated part typically needed a full redesign before it could be printed.

What we did

Defined the upright's load cases, packaging envelope, titanium AM process and a mass-minimization objective at a stiffness target.
Aether generated geometry with FEA in the loop on every candidate — the load path that placed material was the one that validated it — and additive constraints enforced during generation, not after.
Explored thousands of candidates via a surrogate, then confirmed the finalists with high-fidelity nonlinear FEA against the real load cases.
Each result shipped with its margins, the load cases it survived and the print setup it assumed.

In their words

“The old tool drew a shape and hoped the solver agreed. This one generates and validates in the same breath — it printed right the first time.”

— Head of Design, Velocity Motorsport

The outcome

The chosen geometry came in about a third lighter than the incumbent part at the required stiffness, validated against the full load set.
It printed and qualified on the first attempt — no post-DfM redesign loop, because the manufacturing constraints shaped the search.
The Pareto front let the engineers pick the mass-stiffness trade deliberately rather than accept a single black-box answer.
Generative design with simulation in the loop is now the default for mass-critical parts.

One detail worth calling out

The win they hadn't expected was the qualification paperwork writing itself. Because every candidate was validated by the same FEA the result was checked against, the margins and load cases came attached to the geometry — the sign-off evidence was generated with the part instead of reconstructed after.