Exhaust Blown Floor - Forward Exhaust Exit

[ringo]

Shelly i joke in most of my posts.

No i was asking for a reference from you.

[malcolm]

I agree with Shelly: the pressure-scale is too narrow. It's like a stereo where you turn up the volume too far; the amplitudes of the sound waves get cut off, and you end up with a bad representation of the intended sound.

The same thing is happening with your pressure plots... While the CFD might be more accurate than that of Mr. Wirth, if you show a narrow pressure window, we end up with a poor representative image of an otherwise accurate plot. Nawmean?

The minimum you go to is 101000 Pa, and I really down that the last 1.5 metres of the floor is exactly between 100950 and 101049 Pa. The minimum should go to something closer to 100500 or even 100000 Pa, so that we can see what is actually happening at the diffuser mouth, rather than merely just seeing that it is "somewhere lower than 101000 Pa".

PS. Perhaps try to work together to solve the problem ("where does the air go"), rather than putting the problem between you and bickering over who is right. I learned that theory/method in a pre-marriage class...

[shelly]

I think that, being the speed 90 m/s, a sutable range for pressure is from 90000 Pa to 106000 Pa

[PNSD]

Im with Shelly regarding cp's. Just calculate the cp's and that should give a better idea .

[ringo]

The pressure range doesn't affect the calculation.
It's just for display purposes.

This is a good range for the floor. I can work from this and make subtle changes to the geometry.
If it's too wide everything will look either a sea of green, red or blue. You cannot work from that. You want sharp colour contrasts.
A bigger range wont give you that.

I can change the range free of will. I was simply using this range as it was of interest at the time.

In this case, i am not interested in high pressures over atmoshperic. Those will all appear very red. They didn't interest me at that specific time i took the shot.
My interest was with the suction peaks. Which are clearly represented by the data.

This is the same calculation:


Notice the under floor speeds and the range of speeds?

Can it be said the floor is not working?

At this point my interest is not with the diffuser, it's the front splitter and the sides.
Notice that more colurs there give more information.
I can also select any point on a surface and data, but i can't represent that here.

[ringo]

Im with Shelly regarding cp's. Just calculate the cp's and that should give a better idea .

It wont give you a better idea, and you're going to have to pay for that, becuase you are asking for something in 3d which is not the case for most Cop plots.

It would look like this:



This is the first one i did.

And i only used it to figure out the relationship with the COG of the car.

[malcolm]

What you seem to be missing is that Shelly and I believe that the diffuser should affect what happens at the sides of the floor, whereas you seem to disagree; however, the evidence you supply seems to be showing results that are not seen on other plots.

Take the Honda one you showed: forgetting the front splitter, you can see that a streamline makes it all the way out to the edge of the floor, and then gets pulled back into the diffuser. Yours does not show that at all, and the diffuser seems to have much less of an effect.

The plot that you are showing hides some of the data. I agree with you that changing the scale does not change the calculation (obviously), but it SHOWS more of what that calculation just solved. Right now you are just showing a narrow slice of that calculation, and the area around the mouth of the diffuser is just a big blob of blue, since you cut off the colour plot higher than the minimum pressure generated by the diffuser mouth.

[ringo]

What you seem to be missing is that Shelly and I believe that the diffuser should affect what happens at the sides of the floor, whereas you seem to disagree; however, the evidence you supply seems to be showing results that are not seen on other plots.

What plots?

I did not say the diffuser has no effect on the sides of the floor. What i said strictly, is that it cannot "suck" anything like the exhaust due to the nature of the floor design, a T floor.
The Honda plot is not representing anything close to what is being describes as air getting sucked under the sides.

Take the Honda one you showed: forgetting the front splitter, you can see that a streamline makes it all the way out to the edge of the floor, and then gets pulled back into the diffuser. Yours does not show that at all, and the diffuser seems to have much less of an effect.

You can't forget the front splitter.
It didn't get "pulled in" it went to the tyres first, lost energy, then was entrained.
Did it go anywhere in sides though? Since we seem to be conveniently dodging that one.

The plot that you are showing hides some of the data. I agree with you that changing the scale does not change the calculation (obviously), but it SHOWS more of what that calculation just solved.

So you are calling me a liar then?

Trust me i will show anything you want to see, and i will agree with anything that has enough evidence. I don't for being right or wrong.

I am putting to you that no data is being hidden, unlike the colour plots, flow lines cannot be hidden.

Right now you are just showing a narrow slice of that calculation, and the area around the mouth of the diffuser is just a big blob of blue, since you cut off the colour plot higher than the minimum pressure generated by the diffuser mouth.

Oh boy.
You see how they handle me on this forum?
You are being very cynical.

The whole discussion wasn't about what happens in that blue area. And i din't plan anything because that plot was made 3 months ago.

[malcolm]

Don't assume every comment is a personal attack.

What plots? I was referring to the Honda plot, specifically. Also, the streamline I was referring to had a nice, gentle arc; it did not crash into the rear tire and lose energy.

Also, in what I am describing, you can forget the front splitter. In fact, the front splitter gave that Honda streamline a lateral momentum outward, which then gradually gets accelerated inward toward the diffuser. In that case, not only did the diffuser pull the air inward, it generated sufficient suction to pull air inward that was diverging. Given that, if there was a free stream ahead of the diffuser, it would have a wider reach than what is shown in the Honda plot. However, this is all hypothetical, given that the Honda splitter probably wouldn't affect the air much differently than the Renault (there would be differences, but they would be minor and likely not fundamental).

Fact remains, it did not crash into the rear tire.

The streamline I speak of is pointed out by the red arrows. I included a smiley-face as evidence that I am not cynical, but in fact a rather nice guy:




Haha, I am not calling you a liar. I am not saying that you have some premeditated agenda. I am just saying that by cutting off the pressure-scale at 101000 Pa, there is hidden information in the big blue blob around the diffuser. That plot probably served your purposes just fine 20 pages ago for another discussion, but now that we are talking about the effects of the diffuser on the sides of the floor, which is an extension of discussing how the exhaust would likely end up near the sides of the floor (a valid tangent/extension, methinks), your plot hides a bit of information that could be useful to have.

No cynicism, no lies, no accusations, no attacks... I am just pointing out that on part of your plot, the colours don't show all of the information because of the limits you put on the pressure-scale three months ago.

That's a fact, not an insult or an attack.

Can't we all just get along? Group hug?

[ringo]

malcom that flow is not coming from outside the floor.
It entered from the front.

It is that very flow that prevents anything from coming in from the sides.

I want real evidence from anyone here to suggest that flow is coming from outside.
Any evidence at all.

On top of that, someone needs to get that clip of the renault entering the concrete dust cloud.
I know that it can be done, so can someone please capture it?

It's really an end to this discussion. Someone get it for the thread please.


edit: i think you are missing the point of the discussion that part you labeled "BAM" is not what i was arguing. That part is obvious.

The crux of the arguement is " does the diffuser suck in the r31 exhuast which blows at 90 degrees to the car which varies its trajectory as the speed changes"

That is the arugemnt.

[malcolm]

I think they are all pieces to the puzzle.

1) In the Honda case, air goes out to the edge of the floor, and then gets pulled back in.

2) If hot exhaust creates a greater suction peak at the leading edge of the floor of the Renault, those streamlines that would diverge to the edge of the floor on a Honda will be pulled laterally inward as they transition under the floor of the Renault. Now the air that entered from the front doesn't diverge as much.

Given 1 & 2, since the diffusers of each car would be relatively similar in effectiveness, and since the air that entered from the front doesn't diverge as much, it stands to reason that some air from outside the floor would then get pulled in toward the diffuser.

That's my line of reasoning; it may not be the most accurate, but so far I haven't read or seen anything that seems concrete enough to make me change my opinion. Unreasonable? Maybe. Then again, using logic to analyze aerodynamics is just asking for horrid inaccuracies. I don't have much option, however, since I don't have CFD to test my theories.

So in conclusion, the diffuser itself doesn't suck in the exhaust.

What I think happens is that it is the combination of the cross-flow air, the suction-peak of the floor's leading edge, the slightly-lower-than-ambient pressure zone under the whole floor, and the suction-peak at the diffuser mouth all work together to pull at least some of the exhaust ultimately into the diffuser, despite the 90~100 degree angle of the exit.

Just found this... not really conclusive, but shows that exhaust doesn't necessarily shoot out as energetically as perhaps we assumed/calculated on here:

http://www.youtube.com/watch?v=apiB8ia8wos

I don't know the angles of the exhaust exits, and some energy would be removed from the turbos, but given that it exits into a free stream rather than toward a suction peak, and that the speed of the car isn't overly high, maybe that offsets it enough to bring it back into the ballpark?

[shelly]

ringo, pu the joes aside for a moment. malcom has explained weel waht is to be noticed in the honda picture. I think that you understand but you just pretend not to.
The same for the pressure ranges: it is obviuos that with the colorscale you chose for pressure and velocity it is not possible to see if the difffuser is working. Is freestream speed is 90m/s, then you should have pressure from 90000 Pa to see significant suction peaks (cp around -3).
With a suitable color range, more issues would be visible.

The black dots in your pictures are related to the cell centers, aren't they? Which skip factor are they?

[malcolm]

Also, if free-stream speed is 90 m/s, why is the maximum velocity on the scale only 80 m/s?

[ringo]

I said it already. At the time that was done, it was just a random shot.

I still don't agree with any of what you say malcom as it is simply unfounded.

When i get home i'll deal with that point by point.

1) In the Honda case, air goes out to the edge of the floor, and then gets pulled back in.

Yes, exactly, so you are agreeing with my MS paint then?

Now tell me where does outside air come in?

2) If hot exhaust creates a greater suction peak at the leading edge of the floor of the Renault, those streamlines that would diverge to the edge of the floor on a Honda will be pulled laterally inward as they transition under the floor of the Renault. Now the air that entered from the front doesn't diverge as much.

Nope. It diverges even more. Look up the ejector principle.

Given 1 & 2, since the diffusers of each car would be relatively similar in effectiveness, and since the air that entered from the front doesn't diverge as much, it stands to reason that some air from outside the floor would then get pulled in toward the diffuser.

Nope Nope.
It doesn't stand to reason, becuase i have't observed that. I have to see it first.
I respect your opinion, but i have to see it, even if you draw it in MS paint and explain.

So in conclusion, the diffuser itself doesn't suck in the exhaust.

Exactly.

What I think happens is that it is the combination of the cross-flow air, the suction-peak of the floor's leading edge, the slightly-lower-than-ambient pressure zone under the whole floor, and the suction-peak at the diffuser mouth all work together to pull at least some of the exhaust ultimately into the diffuser, despite the 90~100 degree angle of the exit.

Tell you what i will revisit the model and do the CP plot for the mid plane.

[ringo]

http://www.youtube.com/watch?v=apiB8ia8wos

I don't know the angles of the exhaust exits, and some energy would be removed from the turbos, but given that it exits into a free stream rather than toward a suction peak, and that the speed of the car isn't overly high, maybe that offsets it enough to bring it back into the ballpark?

How fast is that engine revving?
It's probably not even doing half that of an F1 car.

and still it doesn't get taken up in the low pressure behind the car.


Imagine 220m/s of hot gas shooting out 90 degrees. Will we expect that to bend and go under?