poniedziałek, 16 kwietnia 2012

How does diffuser work ?

I got a request for this explaination and it became a little bigger, than I thought, so I've decided to post it here.

Here is a model I did for the simulation - yellow part is the ground and blue part is the diffuser-like something.

Basic concept: The air travelling under the car gets squeezed, by decreasing area of the channel (diffuser) – it moves with higher velocity causing the drop of pressure.

To understand this concept we need to analyze two most crucial parts of it:
1. Air moves faster through narrow channel.
2. Higher velocity means lower pressure. 

First point seems pretty logical, as it can be referenced by many examples from everyday life. One of them can be water flow in a river. If the river is deep and wide, water travels slowly. If it’s shallow and narrow water rushes through riverbed quickly. Other examples of this phenomenon are e.g.: flow through shower head, various nozzles in water hoses or sprinklers, gas dusters and many more. Physical explanation comes from conservation of mass principle. It says that total mass of an isolated system must remain constant. Let’s think  about water flow in pipe. The pipe has a diameter and after a while it decreases. Now let’s take two planes in the first part of the pipe. Water between them has certain volume. If this volume is put in the narrower part of the pipe, the distance between boundary planes must increase – because no mass can be created or vanished. Our volume must travel through both parts in the same time - and since the distance between the planes is greater in the narrow channel it must flow faster (that’s because velocity is displacement during a time interval – and if the time is constant, velocity grows with the growth of displacement).

Flow trajectories through diffuser. Notice the faster (red) flow in narrow part.


Now, the second thesis may seem to interfere with common sense. Let’s think again of a pipe with different profiles. We’ve agreed that water flows faster in narrow channel. That would mean that it has lower pressure in narrow channel also. This can be explained using the law of conservation of energy. It says that total energy (in this case of our fluid volume) is always constant. Total energy of flow is sum of kinetic energy, potential energy, and work. Since potential energy depends on change of height above the ground we can rule it out (assuming that our pipe is on the same level from one end to the other, which is true in this case). That leaves us with kinetic energy and work. Kinetic energy depends on velocity and work depends on pressure. That means if we increase one term, we have to decrease the other to keep the sum constant.
Looking from physical point of view laws governing diffuser work are quite simple. And, in fact, a diffuser is just a pipe, that narrows somewhere near the rear axis of a car generating force that sucks the car to the ground. Of course, if one wants to maximize downforce gain, it gets more complex than that. 

Flow trajectories through diffuser. The blue part is the lowest pressure.


 
Disclaimer:

Explanation is quite simplified, so it’s understandable for someone who is not familiar with the mentioned laws and equations. If you seek for deeper knowledge start from googling “bernoulli’s principle”.

Simulation in this case is not really computing anything, these are just pretty pictures showing the general behavior of air in diffuser, so any numbers are irrelevant (well except maybe for a really small race car ;-)) because of the Reynolds number not being preserved.

Also I decided to make this simulation internal, as it is enough to show the Important things, and saves some time, but one has to be aware that diffuser is actually above the ground.

środa, 4 kwietnia 2012

SolidWorks Flow Simulation - F1 frontwing

Lately I've started learning solidworks and flow simulation aswell. Today I've done some simulations on my frontwing, just to check if some of my ideas will work. Unfortunately for me, due to my lack of meshing knowledge it takes a lot more time, than it should and in fact I can't mesh some areas at all. I guess I'll build more simple model anytime soon, so it won't require so much power and it'll be good excercise for me to learn SW. Anyway here are two pictures of flow trajectory, one being velocity-dependent coloured, and the other one pressure-dependent. I've set initial velocity to about 55 m/s which is approx. 200 km per hour. The simulation apperas to be quite correct, however areas of speed difference above and beneath the wing seem to be wierdly large (first picture). I blame it on the fact that there is no ground implemented in this simulation. Pressure figures might look odd too (just 5000 Pa peak-to-peak amplitude), but it's 2D simulation, so they're not really representative. The measured slice generated about 50N downforce (again 2D+no ground).



niedziela, 11 marca 2012

czwartek, 1 marca 2012

Random stuff

Still working on that hurricane. Meanwhile here's a little diffuser-like-thingy I've made using Catia, and a front wing sketch in... paint ;-)





sobota, 25 lutego 2012

Autodesk Inventor Hurricane - part I

Hawker Hurricane - WWII Royal Air Force fighter. I'm intending to play with animation some more, when I finish this, however it's pretty far from being complete. Still not sure how much detail do I want, but it already needs some minor tweaks in tail section and fuselage aswell, so I guess I'll do whatever will satisfy me and will be able to animate. The very next things in my list are ailerons since I'm obviously going to need them ;-).

wtorek, 14 lutego 2012

Autodesk Inventor radial engine

Radial engine made just to show how it works. This is my very first animation and it took me suprisingly long to figure out how it's done in autodesk animation studio. I've added a propeller and a simple connecting rod later.

pictures:
 
and the video:


Autodesk Inventor Shelf

This is something neither aero non mechanical actually, but I've decided to drop it here anyway. It's a project I had to make for school - shelf containing both welded and bolted connections and as much standardized parts as possible. I also had to use specified dimensions and part shapes.


renders:

and some drawings: