Hacker News Reader

Comment by duxup | Hacker News Reader

I've seen wind tunnels in use in videos and TV shows.

Question:

Let's say I had this device and I had a small vehicle and I put it in there to make changes to. Would this tunnel help me actually create a tiny car that legitimately is faster in some measurable way outside the wind tunnel?

I know there's a lot of unknowns in my scenario, but I'm thinking about the utility / outcome of using this device, just in general beyond the cool factor. Is a tiny wind tunnel actually likely to produce real life results for someone?

Edit: Thanks for all the answers everyone, informative info.

View on HN

I don't think it would make any difference for, say, an RC car that could fit inside the tunnel. At the speeds where aerodynamics would make a measurable difference to drag, other aspects such as traction and downforce would be considerably more important. I don't see how it would be plausible to measure downforce with this miniature wind tunnel, because there's no strain gauge to be seen.

For aircraft of any size this would have more utility. This is because, even at low speeds, aircraft must be balanced (center of mass and center of pressure very close to each other) and ideally won't stall at a low AoA (Angle of Attack).

In any case, it would only be practical if you used this with scaled-down models (as used to be the case with real aerodynamics), and it is more expensive and less accurate than doing a computer fluid dynamics simulation. With the simulation, you can do it at full scale in every situation imaginable with high realism and accuracy.

In principle, yes, a wind tunnel would help you create a car that is either faster or more efficient, or to improve the ground effect, or let you control it better. It does that by letting you check the places where the airflow becomes chaotic, that can't be simulated very well.

On practice I doubt you are trying to optimize a tiny car. And the chaotic flow is very dependent of scale. So the only parts that transfer from a tiny car into a normal one are the ones that are easy to simulate.

You can but you are almost certainly going to get better results using a computational fluid dynamics (CFD) feature of your CAD software. This simulates the wind and shows you exactly where and how much pressure is building up. You can iterate on your design without producing a new physical model though it is very compute intensive.

FreeCAD has a plugin to do it and there are videos on YouTube of people showing you how to use it.

Roughly turbulence scales according to

https://en.wikipedia.org/wiki/Reynolds_number

if you adjust things so that the Reynolds number is the same for the model and the real thing.

Somebody in the next building over had a "wind tunnel" that used this stuff

https://en.wikipedia.org/wiki/Sulfur_hexafluoride

which is 5 times denser than air and has a higher Reynolds number. Oddly SF6 is safe to breathe (as an inert filler gas such as Nitrogen, Argon or Helium) but is also one of the most powerful known GHGs.

Wind tunnels for scale size cars are much less useful than they were just 15 years ago. My university sold its wind tunnel for 1/10th scale cars, it was just not that useful compared to CFD. CFD can get you simulations that take into account rotating tires (a big deal on the open wheel cars we were working on). You also don't have to worry about inaccuracies from 3D printing (or however you make the model).

You need to make sure your measurements are actually representative of the flow regime you will actually have. You need to look at Reynolds number and mach number. If you use the actual model and don't use a small scale model, you just need to use the same flow speed as you will use in the application later. Then you could theoretically measure drag force and look at vortices to check where to modify the shape.

Reply on news.ycombinator.com