Rear Diffuser Issues

[slimjim8201]

Folks,

Looking to gather insight and knowledge from the technical community regarding rear diffusers. I've conducted numerous CFD studies over the years which have consistently shown the adverse interaction between the rear wheels and the low pressures generated by the diffuser. The challenge I've noticed is that air is generally pulled towards the centerline of the car and the outermost portions of a diffuser are not utilized. Here are a few images highlighting the phenomenon:









Obviously, these images show relatively straight, un-contoured diffuser sections and strakes, but I've seen similar effects across a very wide range of diffuser sizes, positions, angles, approach (convex versus concave). Even through examining modern supercar and LeMans prototype style underbody/diffuser designs, this phenomenon appears consistent. Would anyone care to share their experience?

[Pierce89]

If you look at cfd of some very high level race cars, you'll still see that phenomenon. The under body flow always seems to converge at the low pressure zone. The f1 ebd's used a high speed vortex to push the wheel wake away from the diffuser. If that's not possible, some large strakes could help.

[slimjim8201]

I'll post some more pictures of my latest testing examples, but I think I'm on to something.

The position of the start of the diffuser may be the culprit. If placed to far forward, the lowest pressure region occurs at or ahead of the rear tires. Air then travels from the high pressure zone in front of the tires laterally towards the center of the car. If the diffuser entrance is placed further back, at the rear axle line for instance, the low pressure zone occurs in between the rear wheels and the air must travel longitudinally from high pressure to low pressure.

[P.S.]

Hi slimjim

Your investigation about the airflow is moving towards the centerline is a fact and probably hard to challenge.

The more general target you have, is to create downforce with the so called ground effect. And when you have no skirts, like it was the design in F1 in the eighties, a low pressure zone under the car will always suck air from the higher pressure region from the outer sides. As you say this effect is the same in all your investigation. And I say it will always be.

The other more accessible challenge is to get your diffuser more efficient (which will create an even higher contraction of the airflow from the outer sides and improves the downforce). To reach this you like to have the airflow more attached in the diffuser section, less separation. This optimisation is not so easy with CFD in general and especially with such a rough mash as I see on your pics. I´m absolutely no CFD guy but this is all about boundary layer and there you need lots of computing hard ware power and experiance... Have you ever tried to compare your results with reality?

[olefud]

Keep in mind that a “diffuser’s” purpose is to slow down the air from the higher relative velocity caused by the vehicle displacing and channeling air. The thought is to regain pressure so that high speed air at the rear of the vehicle does not produce a retarding low pressure volume behind the car. Not to way that a pinch area under the vehicle might not develop a low pressure, high velocity area.

[PlatinumZealot]

@ OP.

You gotta do it at different speeds. At high speed there is a lot of turbulence behind the wheels and the program won't show this in detail either because the overall velocity here is very low or your mesh is not detailed enough.

[olefud]

@ OP.

You gotta do it at different speeds. At high speed there is a lot of turbulence behind the wheels and the program won't show this in detail either because the overall velocity here is very low or your mesh is not detailed enough.

Or more;

http://mccabism.blogspot.com/2014/08/ad ... y-cfd.html

[tuj]

It's also important to note that the F1 teams are 'stalling' the diffuser in various ways (according to Racecar Engineering), and that this is critical to minimizing drag.

[trinidefender]

The stalling diffuser idea is dubious at best and I would almost lay my head on a block to say that while some stalling of the diffuser may happen at the very highest of speeds, that it isn't a primary concern to the designers. At a point when I have more time I'll explain my reasoning.

[tuj]

Scarbs was talking about stalling diffusers 7 years ago. I can't believe he's that far off.

http://www.autosport.com/news/report.php/id/57511

[Cold Fussion]

The idea of the diffuser stalling to reduce drag seems pretty dodgy, I would have thought a well designed diffuser would carry a small drag penalty.

[tuj]

My understanding, albeit limited, is that yes, certainly the diffuser produces less drag than a wing, but it *still* creates drag. This is going to be true of I think any downforce-generating device; induced drag comes along with -L. Yes, I think the diffuser has less drag than a big 'ol rear wing.

Again, my limited understanding is that the height of the diffuser is critical as is the rake of the car. Too high and you will suck in air from the tires, which is turbulent and bad. This is why the teams used the 'exhaust-sealed' diffuser design, particularly RBR. Go too high and the flow will also detach and create turbulent air behind the car rather than a smooth(er) wake. Let the diffuser get too low to the ground and you will choke off flow and lose the downforce.

In this latter way, the diffuser can be thought of as 'self-stalling' at high speeds if the car is not stiffly sprung. As speed increases, the total aero load from the aero devices on the car increases. This pushes the car closer to the ground, reducing the effectiveness of the diffuser. But let's say we have a stiffly sprung car, like an F1 car, where ride height is 'relatively' constant and the mechanical suspension is more or less in the sidewalls of the tires; now the diffuser is effective at high speeds.

But this is generally bad, because usually when we are traveling in a race car at high speeds, we are on the straights. So the idea is to stall the diffuser on the straights either by running it too low to the ground or by some other means. But the other problem is, we need to 'unstall' the diffuser when it's time to brake, which will also likely be at a high speed. I think this is one of the major reasons FRICS was being used for the last 6 or so years in various forms; it allows for better rake control under all dynamic aspects of the chassis. It also, if done right, allows the diffuser to stall but then regain airflow for the braking phase.

Again, I am no expert like Scarbs, but if Ferrari was stalling the diffuser with a moveable floor 7 years ago, then 1) stalling the diffuser is apparently worth a large engineering effort and 2) I'm sure everyone in the paddock is looking for ways to implement the same tricks.

Now you may ask "how does this work without FRICS?" I honestly don't fully know, but based on the Mercedes pace at and after Germany when the suspension devices were removed, I think Merc always had an alternate and equally or nearly equally effective plan.

It's all about -L/D in the whole package. Watch Formula Student and you will see a lot of 'draggy' wings because their top speeds are low. Watch LMP cars at Le Mans and top speed and reducing drag becomes very important.

Funny thing is, I asked the exact same question last year when we were talking about stalling diffusers: http://www.f1technical.net/forum/viewto ... =6&t=15480

Just my humble thoughts.

[Pierce89]

The stalling diffuser idea is dubious at best and I would almost lay my head on a block to say that while some stalling of the diffuser may happen at the very highest of speeds, that it isn't a primary concern to the designers. At a point when I have more time I'll explain my reasoning.

I understood it to be not "stalling" per se, but has the flow choked off from viscous blockage at very low ride heights.