Influence of vortices on underfloor aerodynamics

[McMrocks]

What we see in the caterham video is the vortex generated from the barge boards

As opposed to...?

If I've implied a different interaction, it's purely unintentional.

I wasn't disagreeing to you. And would like to apologize if you understood me like that. I just thought the picture is helpfull as it clearly shows the vortex underneath the floor.

However i would really like to see a CFD from a current F1 car in that area. The current bargeboards are WAY larger than that in the Renault( or McLaren) in that picture. The vortex generated by them must be even more powerful.

Before that discussion i thought that the barge boards don't generate a vortex. I thought they are in line with the curved airflow around the sidepods. Now i understand that they are not only but also a big vortex generator

[hollus]

I think the flo-vis leaves the sidepod due to the overlap of an overall flow which would follow along the sidepod, with vortex rotation (the Y250 vortex) like the blue vortex in the illustration below. The net effect is a significantly downward flow in the car's skin and diagonal lines in the floor. The flow that hugs the rest of the sidepod originally was in the other end of the vortex and did not pick up any flo-vis paint.

[bhall II]

I wasn't disagreeing to you. And would like to apologize if you understood me like that. I just thought the picture is helpfull as it clearly shows the vortex underneath the floor.

However i would really like to see a CFD from a current F1 car in that area. The current bargeboards are WAY larger than that in the Renault( or McLaren) in that picture. The vortex generated by them must be even more powerful.

Confused apparently am easily I sometimes.

If you compare the current generation of barge boards to those that came before, you'll see they've actually become quite puny...



When the Overtaking Working Group recommended substantial downforce reductions for the 2009 season, they took most of it from the barge board area. Prior to that, teams employed absurdly vast arrays of vortex generators and other flow conditioners...



In any case, I think this is what's happening along the outside edge of the floor here...



The streamlines on the barge board indicate that it's guiding air flow directly under the floor. Those on the outside of the flip-up show that (at least some) air flow from under the floor merges with air flow over the floor to create a counter-clockwise vortex that would seem to travel aft along the sidepod. Such rotation would tend to seal the top of the floor downstream, preventing air flow from being sucked under and degrading efficiency. Flow from behind the barge board that appears to be discarded could actually be rolled up into that vortex.

Whether or not the "barge board vortex" and the "flip-up vortex" are actually one and the same, I don't know. Maybe, maybe not, maybe sometimes. It's difficult to read, because conditions aren't static; they change, especially when cornering.

[chuckdanny]

Bhall ? are you angry ?
I'm angry to disagree on your interpretation of those flowvis. For me the vortex turns exactly the other way around.
Why? because the flip up slows the airflow so increase pressure, air tends to flow from high to low pressure (a very famous aphorism from you) which is the underside side. The vortex arise from the corner discontinuity, the horn shape which is not without remembering the angry bird wing. What is the low pressure side on the angry bird is pretty straightforward to visualize. The flip up blows higher pressure air tangentially to reinforce this vortex, it would brake if it was turning the other way.
Last but not least, the flip-up flips down and outward at the most afterward tip to counteract a tendency of the vortex to shift upward following the inflow pattern of the cock bottle shape.
We can clearly see on those flowvis that the helicity at the beginning shows the vortex rotation... but it changes at the afterward position where the vortex bends down. this part is misleading showing a reversed helicity which in fact shows flow separation.
I may add that the sidepod undercut outwash, further reinforce and keep it down.

I think you need strong opposition to give the best of yourself !

[bhall II]

Not saying you're right or wrong. Just try to visualize the car in right turn...





(Yes, I know the animation is a left turn. But, it took me forever to figure out how to make it, and I'm not doing it again. Ever.)

[trinidefender]

From some models I have seen it shows the vortex rotating counter clockwise on the right side of the car and clockwise on the left side of the car (looking from the back of the car forward). Don't have much time now but I'll try to do a write up with supporting evidence at some point soon.

[chuckdanny]

We can clearly see it on the renault cfd, unless you explain that for some other reasons it could this year be otherwise.



So, without further evidence, it is counterclockwise on the left as seen from the back.

[bhall II]

You also see very weak vorticity, both in the CFD and the FloVis streamlines. So, it's not altogether clear that conditions are static. Again, try to visualize a right turn.

Counter-clockwise rotation (from the front) matches the Y250, and it has better sealing effects above and below the floor. Thus, it seems logical that such a flow structure might be an active target for designers.

[henra]

From some models I have seen it shows the vortex rotating counter clockwise on the right side of the car and clockwise on the left side of the car (looking from the back of the car forward). Don't have much time now but I'll try to do a write up with supporting evidence at some point soon.

I tend to agree. Looking at the shape of a typical barge board this seems to make sense. The flow would start 'rolling' over the TE of the barge board at the front and this would create a clockwise Rotation on the left side of the car when seen from the rear . The positioning of the lateral 'diffuser' and the guerney seems also to support that. You can see that it is positioned rather forward and there is a clear gap in front of the sidepod. This would obviously counter any anti cloickwise vortex in that area. The Y250 is also clockwise on the left side of the car, when seen from the rear if I'm not completely mistaken. A clokwise Rotation will also better protect it from entering underneath the floor along the sides of the sidepod.

[bhall II]

Just to make sure we aren't describing the same thing from different perspectives, let's standardize our point of view, shall we?

Facing aft, thus downstream, the red vortex is counter-clockwise, and the blue vortex is clockwise.



Does that work for everyone?

[PlatinumZealot]

Just to make sure we aren't describing the same thing from different perspectives, let's standardize our point of view, shall we?

Facing aft, thus downstream, the red vortex is counter-clockwise, and the blue vortex is clockwise.

http://i.imgur.com/4wroVgu.png

Does that work for everyone?

I agree with the blue vortex. The red one won't happen as you draw it.

[chuckdanny]

The horn vortex is like a wing tip vortex, i don't think that you can change its sens of rotation, it is the child of pressure difference between floor top face and underfloor suction(no sexual reference here unlike your draw wing ).

And there could be a clockwise one in between the bargeboard and the floor tip one preventing a merge so the horn one from flowing into the diffuser. The aforementioned being triggered by the horn/floor leading edge discontinuity.
Here :

[bhall II]

(no sexual reference here unlike your draw wing ).

Perhaps we see what we want to see?

...cock bottle shape.

Not judging.

I agree with the blue vortex. The red one won't happen as you draw it.

I think my mistake was emphasizing vorticity.

First, let me preface this by saying the following is based upon the idea that air flow under the car is bad and should be avoided wherever practicable, because the rules don't allow for aerodynamic elements in that area, such as proper venturi tunnels. Instead, the idea is to create suction by sealing off the floor with vortices that mimic the effect of the old sliding skirts.



In practice, only the (EDIT: p̶l̶a̶n̶k̶) reference plane can be truly sealed since the rules require a step plane.



I don't know if this idea is accepted as conventional wisdom or not. For instance, I'm not always sure what's meant to be conveyed when someone says high noses attempted to cram mass flow under the car. To enhance sealing vortices? Fine. For anything else? No. The point being: if anyone disagrees with this idea, I think that's probably a conversation for another thread.

At any rate, I think what we're seeing below is the result of vented air from under the floor.



Air flow follows a parabolic trajectory as it travels from the front edge of the floor to the diffuser, and its velocity is inversely proportional to its distance from both. In other words, air flow slows down at the middle of the floor, which means the middle of the floor loses efficiency. The solution is to get rid of such flow altogether, because it's definitely unneeded.



How? With floor vents (flip-ups, little diffusers, whatever). The images should be self-explanatory. (Note: the curvature of the arrows is not meant to imply a specific flow structure.)


The MP4-24's method was weird and indicative of absolutely terrible air flow around the Coke-bottle area.



It seems this effect is enhanced by placing a vent near the floor's leading edge. As PZ mentioned earlier, the additional outboard acceleration -- clearly seen below on the FW31 as reduced pressure -- should bring the floor's center of pressure forward.


FW31 pressure coefficients

If so, the vortices will form a "soft" seal along the front edge of the floor.


Red: bargeboard vortex
Blue: other flow
Everything: highly generalized and exaggerated

I think it's likely this behavior is most pronounced in yaw before the streamlines can "catch up" to the car, so to speak. It's also possible that yaw angle is what initiates it to begin with.



The usual caveats apply: this is just how I see it; I could be wrong; blah, blah, blah, blah...

[chuckdanny]

I know you are discarding everything i say because i say it but it won't stop me from giving my point of view.
What happens is that even if it is not a smooth unmistaken venturi it is a kind of !
And for an (open) venturi, the convergent part is what create this outwash, its the third principle of dynamic action/reaction, being forced hence pressurised to accelerate it either effectively accelerate or expands sideways where it can that is at the floor edge to begin with.
In the divergent part its the opposite hence the change of "gravity force", the low pressure appearing at the throat and extending sucks air from the sides also and of course the diffuser.
So you seal the floor for different reasons at the convergent or at the divergent.
And i wil say that my exotic research into a parallel world may have given me the answer to why maybe you shouldn't completly seal the divergent part because you can massively benefit from vortex enhanced suction.

Air flow follows a parabolic trajectory as it travels from the front edge of the floor to the diffuser, and its velocity is inversely proportional to its distance from both. In other words, air flow slows down at the middle of the floor, which means the middle of the floor loses efficiency. The solution is to get rid of such flow altogether, because it's definitely unneeded.

It's true, i had exactly the same pattern on a simplified blablabla. The adverse pressure of the rear tire is also a reason.
The horn vortex turning as i see it with core under the floor level washes out the high pressure building from the floor wall while it still allow inflow under the vortex core.

[bhall II]

I know you are discarding everything i say because i say it but it won't stop me from giving my point of view.

I don't discard everything you say just because you've said it. Frankly, a lot of the time I can't make out what it is you're trying to say. It's not your fault, but the language barrier is intense.

Anyway, you can't completely seal the floor, if for no other reason than it would stall the diffuser. That's why I said air flow under the car is to be avoided "wherever practicable."



Air flow through that channel, toward the diffuser, is constant.