I bought a textbook on fluid mechanics recently, having a rough plan to learn more about weather patterns by writing a simulation. Expecting a theoretical physics text, I was a bit unsettled to read the following on the first page of the introduction:

“The theory is often frustrating”

“viscocity .. gives rise, at frustratingly small velocities .. to turbulence”

“The theory of turbulent flow is crude .. yet it can be quite serviceable as an engineering estimate”

Crude? Frustrating? “Engineering estimate”? These are not words I’ve ever read in a book about theory before. I’m a bit scared!

5 replies on “Turbulence”

Weather modeling is very simple: There’s a 70% chance that tomorrow’s weather will be the same as today’s 😉

Which book did you get? The first thing to note about fluid dynamics is that it’s characterised by endless empirical non-dimensional numbers… did you get some time on the T3E? 🙂

It’s by Frank White –

I once went to a Ed Scifest talk by a F1 aero bod. I wanted to ask him to what degree they trust their CFD simulations – as in, do they ever get to the track or wind tunnel and find the aerodynamics are substantially different from those out of the sim. Unfortunately, the chairpersondude didn’t pick me for a question.

Hey, Tim – did you ever compare CFD results with your funky wind tunnel? Presumably there’s a reason why F1 teams and Boeing/etc still have wind tunnels and don’t totally rely on sims?

CFD comparisons: in a nutshell that’s why windtunnels still exist; while CFD is very good (and improving) there are lots of areas (usually the most interesting ones) where the CFD codes just aren’t good enough: hypersonics, anything involving heat transfer & fluid flows, turbulence (in general), simulations where you need to model across a broad spatial scale range, etc. Most of our wind tunnels campaigns were for other aerospace companies e.g. boeing, airbus, eurocopter and in turn we were just a small part of the experimental picture; we could di instantaneous 2D with 2C, 2.5C velocity measurements but not time resolved; LDA is normally used for time resolved measurements but that in turn is point wise…

Comments are closed.