CFD Fire Modeling: The Inverse Problem Nobody Wants to Solve

Fire safety simulation has a problem that’s easier to describe than to solve: the CFD model needs material properties — thermal conductivity, specific heat, ignition temperature, char oxidation rates — but these properties aren’t available in standard databases for most real materials, and direct measurement is expensive and often impractical. The alternative is to go backwards. You have experimental fire test data. You have a CFD model. Find the material properties that make the model match the experiment. ...

Jan 2024 · 3 min · Vishal Sharma

Engine Mounts: The Art of Designing Something to Fail Softly

Engine mounts have one job: let the engine move enough to not destroy itself or the chassis, while not letting it move enough to shake the cabin apart. Those two requirements are in direct opposition. Get it wrong in one direction, and your engine torque reaction cracks the frame. Get it wrong in the other, and the idle vibration transmits straight to the steering wheel and seat. The design space is narrow, the loading is multi-directional, and the rubber changes its behavior depending on temperature, amplitude, and frequency. It’s not a bolt-sizing problem. It’s a dynamics problem. ...

Jan 2024 · 4 min · Vishal Sharma

Front Fender Drag Optimization: Why Running CFD in a Loop Is a Bad Idea

The naive approach to aerodynamic optimization is obvious: run CFD, check drag, adjust geometry, run CFD again. Repeat until satisfied. This works if you have a few design variables and a lot of patience. It doesn’t work if you have a high-dimensional design space, a CFD solver that takes hours per run, and an optimization algorithm that wants to query thousands of points. The solution is to decouple the two problems. ...

Jan 2024 · 3 min · Vishal Sharma

Magnesium Alloy Wheel: What Happens When You Change One Thing and Everything Changes

Everyone wants lighter wheels. Lighter unsprung mass improves handling response, reduces suspension load, and — on a two-wheeler where gyroscopic effects matter — changes the steering feel in ways that are hard to model and easy to notice. Magnesium is 35% lighter than aluminium. The trade is real: magnesium corrodes badly, costs more, and behaves differently under fatigue loading. The project was to find out whether the trade is worth it, under what conditions, and what the design has to look like to make it work. ...

Jan 2024 · 3 min · Vishal Sharma