Exhaust Blown Floor - Forward Exhaust Exit

[horse]

So you can't act like i made a big change.
All i did was drop the pipe a few cm. Be fair.

I hope you don't consider me unfair, but you were advocating that the flow go around the side pods, in principle, for some time. In fact, I'm sure the solution lies somewhere between the two extremes of your tested geometries and I still think you should be commended for putting it in the effort to do the simulations in the first place and further the discussion.

For me, the exhaust initially is directed to round the leading edge of the floor, but then there is still some uncertainty about it's path beyond this point. I was wondering, Ringo, if perhaps your diffuser model might not be operating properly which is why the flow is not being drawn in towards it?

[ringo]

It is not drawn to the diffuser at all. It's the pressure gradient infront of the tyre that deflects it there after a certain speed.

There is no drawing or sucking in. Without the wheel it would blow wide; then maybe come in later, i have to see it first.
The only difference with the car was the leading edge of the floor, that's it. Everything else stays the same.
The logic is intact. The floor does not see most of the exhaust before it rams into the rear tyre.

There is no 100% correct answer, yet. As we see things change with time.

What is wrong though is that it stays under the floor and barrels down to the diffuser and doesn't go the edge.

Since we have better pics of the exhuast i can run the model again, though it will take awhile and i don't really enjoy modeling as much as Legendary M and the others who will spend loads of time on it.
I'll take away the rear wheel and show what happens too.

[horse]

Since we have better pics of the exhuast i can run the model again, though it will take awhile and i don't really enjoy modeling as much as Legendary M and the others who will spend loads of time on it.
I'll take away the rear wheel and show what happens too.

That would be brilliant, if you can find the time/enthusiasm. Cheers.

[shelly]

Exhaust blows horizontally, and then pressure gradients make the jet deflect.

The pressure gradient is actually the static pressure before the rear wheel. That's the plain fact.



see the green and red?

It's all there, posted ages ago.

That rear wheel causes the gas to deflect to the inside.

@ringo: the pressure gradient I am referring to is the one on the leading edge of the sidepod (high pressure on stagnation wall, lower pressure below): this gradient makes exahust go under the floor.

Furthermore, it is not only about rear wheel blockage: air is drawn inside by suction peak on the diffuser kink line, which is one of the biggest overall downforce contibutors.

Finally, the trajectory highlighted in red in scarbs' picture is roughly the same that air would have under a normal (i.e. not fee) floor, with streamtube growing largerafter leading edge sucion peak and the stretching and narrowing by combined effect of wheel blockage and kink line suction.

[shelly]

I have no issue with the initial zone approximately 5D outside the exit.

@raptor22: I do not want to wind you up; I have waited for your definition and exaplaination instead.
I find your posts a bit contradictory.

First you say exhausts are no jet, that do not keep coeherence after exiting the pipe; then you say that behave like a jet for 5D but then do not, because they mix (but also jets mix after 5D...). Then you say that there is no supersonic speed; but it is not relevant to jet definition.
Finally you say that there is no big change in density: but that density is three times (ok, not 1000 times)different is maybe the only thing on which both marekk and ringo agree and neither this is relevant on exhausts being a jet.

[ringo]

Exhaust blows horizontally, and then pressure gradients make the jet deflect.

The pressure gradient is actually the static pressure before the rear wheel. That's the plain fact.



see the green and red?

It's all there, posted ages ago.

That rear wheel causes the gas to deflect to the inside.

@ringo: the pressure gradient I am referring to is the one on the leading edge of the sidepod (high pressure on stagnation wall, lower pressure below): this gradient makes exahust go under the floor.

Furthermore, it is not only about rear wheel blockage: air is drawn inside by suction peak on the diffuser kink line, which is one of the biggest overall downforce contibutors.

Finally, the trajectory highlighted in red in scarbs' picture is roughly the same that air would have under a normal (i.e. not fee) floor, with streamtube growing largerafter leading edge sucion peak and the stretching and narrowing by combined effect of wheel blockage and kink line suction.

The trajectory highlighted is typical yes.
Others didn't buy it all at the time. They were talking about jets and such.
The drawing is pretty accurate, i suppose it was conjured from some source, maybe a cfd image in this forum , because these things cannot be imagined, they have to be observed. If not a CFD image this drawing had to be guided by someone in the know, for the first one with it going directly bellow was not accurate.

Here is an interesting shot though:


Notice the spray near the floor. I am not putting a motion to you, just giving you something to mull.
I will try another CFD this time with more knowledge of the Renault floor.
Whatever happens happens.

[ringo]

alright, turning a new leaf. No more bickering.

just serene pictures:





close enough ?

check the pipe angle and size, also the floor's edge. good to go?

[shelly]

@ringo: one advice: try exapnding the upper and lower limit of your cfd colormap so that it does not get saturated and shows better the results of the simulation.
From the image you posted it is not possible to undertand if the diffuser is working or it is not, like horse suggested

[godlameroso]

I'd love to see the CFD I predict that some flow will still go over the floor.

[ringo]

Ok
i made this before you said that, but i'll post it.


I didn't get the velocity key in, but i'll move it for the others.

notes:

the FEE is sensitive to the bargeboard, the reference plane shape.
It is also sensitive to the floor airfoil at the front.

This result was not to be expected. SO it can be said that tuning will get the desired result.
No one thing is right or wrong.

In this case most of the flow is below the floor. This is because of the pipe blowing under the airfoil.
However it does not go to the diffuser.

Reason being, the ref. plane shape is straight and becuase of the barge board. little more on that later.

Notice the big space between bargeboard and tub for the r31:


i'll make those changes.

[marekk]

@ringo - i really appreciate your enthusiasm for cfd, but without very exact modelling of the car it will produce just nice looking pictures.

As an example - look what a simple change of airfoil's angle of attack can do to the flow:

[ringo]

I changed the airfoil to something similar to renault. I can't be bothered to post so maybe tomorrow.

I know the exact thing is required, but i am at least being enlightened as to what specific aspects affects the flow.

What it also does show is that the human mind cannot predict what is happening either. The under floor is very critical here as well as the splitter area.

No one is capable of relating all the effects of the other parts on the flow of the exhaust. It's best to see what smoke the renualt will blow in a light rain or at the start if the engines are topped up with oil.

This model is more accurate than the one before. But the change in the splitter i made is what is making it diverge even more from the centre. The part of the floor it is blowing under is also affecting this as well.

I figure depending on what Renault want, they can proportion what goes on top of bellow, and what angle and at what speed. It's completely variable.

[strad]

This channel creates a vortex down the side of the car the will act as a skirt discouraging air from migrating under the car. I believe that Renault are using the exaust flow to augment this vortex and enhance underwing sealing.

You hit this one right on the head. =D>

[manchild]

This channel creates a vortex down the side of the car the will act as a skirt discouraging air from migrating under the car. I believe that Renault are using the exaust flow to augment this vortex and enhance underwing sealing.

You hit this one right on the head. =D>

PostPosted: Tue Feb 01, 2011 3:07 am

Perhaps they use exhaust(s) positioned at the front as virtual skirts?

I mean, since volume and speed and exhaust gasses is much faster than speed of car, perhaps the gasses could prevent spilling of airstream sideways? Having them spilled sideways would be no loss since they need to go somewhere anyway, and if they could at least partially seal the space between stepped floor and the longitudinal outline of the floor (at least at the beginning of the floor), than the air traveling below the floor would be less spilled, right?

I mean like these small vertical fins on top of airplane wing that serve for same purpose, only "virtual" since floor must be skirt-less, flat that is.

http://img338.imageshack.us/img338/6938/mig170fn7.jpg

viewtopic.php?f=12&t=9187&p=220920&hili ... rt#p220920

[Ciro Pabón]

Wow...