Under floor flow & diffusers

[chuckdanny]

Thanks! It's the velocity's magnitude regardless of the speed direction, in the wake area of tires for example air is swirling, there are two counter rotating swirl.
I know in reality there are many more of different scales, it's a low resolution

[chuckdanny]

I tested to be sure the vortices reduce static pressure assomption

And it is beautiful ! I've placed a plate above my angry bird :



It's seems to work as expected... and no ground effect which should improve the phenomenon

[chuckdanny]

And now, why not, the 2d model with strakes (not sure it is called strake) at the back !

Surprise, even more thrust and this time thrust appear right where rear wheels winglet are placed.

[chuckdanny]

End now, with a 2d diffuser, i think we've got a winner in a 2d world



Static pressures as advised by CBeck113

Without strakes and diffuser


With

[MadMatt]

I'm sorry but I fail to see the interest in all this without more details from you on your model. Ok velocity contours on a plane close to the ground can help to see what the flow is doing, but you need to watch the pressure distribution on the body. Could you also show the whole geometry for clarity, because it is unclear (to me) what the diffuser or the body shape looks like.

[chuckdanny]

I'm sorry but I fail to see the interest in all this without more details from you on your model. Ok velocity contours on a plane close to the ground can help to see what the flow is doing, but you need to watch the pressure distribution on the body. Could you also show the whole geometry for clarity, because it is unclear (to me) what the diffuser or the body shape looks like.

Well i think basic principle is to accelerate air under the car, benefit from dynamic pressure over the car but not decelerate to much the air that passes over the diffuser because there must be a kind of matching between stream over and under and it seems that speeding the air over the diffuser is a requirement regarding what they do like exhaust blowing or air acceleration over the sidepod and further acceleration when it is falling down to this area.
My 2d diffuser is of course a joke i wanted to see if i could reduce static pressure behind rear wheels.

It's a 2d model only and i don't pretend to get quantitative results with my core2duo even if i succeed one day in using mpich to get the second core processing...

[shelly]

In a 2d model only you cannot appreciate where the downforce of an underfloor comes from - i.e. from the cornering of stream lines in the vertical plane around the leading edge and the kink line.

Wherever there is a curved streamline there is a pressure gradient with pressure going form lower inside the curve to higher outside the curve.

[chuckdanny]

of course, i know that... did i say something meaning that ?

[chuckdanny]

The rear aero of an f1 car is a diffuser plus a so called rear wing main plane with it's flap. All 3 are working together in sync to maximise downforce.
The main plane work like an extrados upside down, flap thanks to the intrados of the main plane that prevent an inevitable stall at such an angle (plus mclaren teeth leading edge) plus dynamic pressure which is also the case for the main plane.

The greater the speed over, under etc... those elements the more work they achieve in a given time. So air is accelerated "naturally" by the wing and the underfloor it's the basic principle of there working. So in order to keep all this in sync you must accelerate air over the floor also... hence the shape of the sidepod bottom close to the floor while it serve other purposes also like sidewash to prevent front tires wake intrusion over the floor. But i showed (maybe) that without the tires themself this speed gain around the sidepod would soon have been lost hence rear tires have a real aero fonction.
But it's not all, the shaping of sidepod as seen from a vertical longitudinal plane contribute to a downwash at the rear (z altitude exchange for v speed in a bernouilli model. Sidepod upper surface faces the sky right? so why would you accelerate like the extrados of a wing to create lift? For the back end...
it's the big picture of a simple there is a diffuser explanation...

[chuckdanny]

Furthermore you didn't ask questions and i gave you answers ! What's the problem ?

[chuckdanny]

A little bit refine with a rounded leading edge for the plate, to correlate let's say qualitatively with the williams render

[chuckdanny]

I don't want to flood(nice one) these forum but i like those cfd render it's spectacular !

[shelly]

height variation of the sidepod does not count in potential vs kinetice bernoulli exchange in this case - if tou check with the numbers you should see that the contirbution is very small at these speed and heights

The picture show nicely the effect of vortices running below a flat surface; to get closer to what happens at the leading edge of the floo, you could lower the flat surface a bit, so that the vortex tube is forced to go around the laeidng edge more, instead of running below it relatively undisturbed.

I the vortex is forced to take a curved path, you shoul see id getting narrower and the low pressure inside it going even lower

[chuckdanny]

Great reply! I have been lazy on this one... didn't evaluate you're right!

The plate was lower initially and there was a big interaction between the batwing and the plate so that a vertical swirl was forming and i couldn't get the solver to converge. So i've been cautious but i think it was more relative to the plate being to close to the wake area of the bird, now that i have a better initial flow field i will try

[chuckdanny]

Job done!

The vortex work so hard that it burst ! As you can see my angry bird is not happy at all