Khamsin Virtual Racecar Challenge 2015

[CAEdevice]

The ratio between DF and DRAG looks more realistic now, but it seems you have to work on it

[RicME85]

Yeah, I need more AoA on my airfoils for a start.

[cdsavage]

Some rule clarifications for round 2. I will be updating the rulebook PDF where necessary (no changes made to the guide file).

- There were a couple of entries which were exported with a very fine quality setting, and this caused some problems. I would ask that everyone try to keep the total size of the STL files (before compression) below about 80mb for ASCII STL files, and below 20mb for binary. This applies mainly to those exporting from solid modelling packages.

- In rule K3.6 (mirrors): the phrase 'with the rear wing removed' refers to the rear wing profiles only, not the endplates. This means for some entries, the mirrors will need to be moved slightly inboard.

- Rule K4.4 (exhausts): Previously we specified a circle of diameter 80mm. We now ask for a circle with area >= 5000mm^2. With very coarse tessellation at the exhaust, there was a small area reduction while still arguably meeting the 80mm diameter.

- Some more wording added to the rulebook for the 10mm thickness rule - all in line with what's been communicated so far in this thread or in emails/PMs.

- For rule K2.4 (wheel visibility), the 'wheels' are considered to be the entire volume of the wheel well (see image), not the wheels included in the CFD parts file. The exceptions (eg 'more than 950mm from the centerline of the car') previously only applied to the front and not the top - they now apply to both.


- Extra wording at the end of K2.6 concerning front suspension mounting: If the front suspension was offset from the footbox in the Z-axis, it was possible to have the inside of the front suspension members hanging in space. This rule now states that there must be a realistic load path to the chassis, consisting of bodywork at least 25mm thick.

- Extra wording in K3.2: it is legal to not use any rear wing profiles.

I would also like to ask for any feedback on possibly replacing the side impact structures with an internal 2d template, parallel to the inlet surface and a certain distance rearward of it, of any shape (with a minimum width and height, and area something like 150K mm^2), which must be contained inside the bodywork. If we were to use something like this, we would not make the change any earlier than round 4, when we switch to the lower-drag rounds.

Now time for some more images

CAEdevice:


JJR Racing:


Variante:

[CAEdevice]

About the side impact structure rule change proposal: I would agree (4th race or next race is ok, better if next race)

About the images: pressure or velocity? I'd prefer to see on the forum a picture of me naked than to see my diffuser sectioned
... but it's ok

[CAEdevice]

hybrid diffuser for everyone: what is your interpretation of the pictures?

[variante]

mmmh... i need time to understand why those pictures show those distributions...
Indeed, pressure or velocity? What program did you use, Chris?

hybrid diffuser for everyone: what is your interpretation of the pictures?

yours is hybrid the other way around

[CAEdevice]

mmmh... i need time to understand why those pictures show those distributions...
Indeed, pressure or velocity? What program did you use, Chris?

hybrid diffuser for everyone: what is your interpretation of the pictures?

yours is hybrid the other way around

I'm working on a new diffuser with two concavity changes! But the biggest issue is the lack of df on the rear wing... less than 250N @100mph!

[cdsavage]

mmmh... i need time to understand why those pictures show those distributions...
Indeed, pressure or velocity? What program did you use, Chris?

hybrid diffuser for everyone: what is your interpretation of the pictures?

yours is hybrid the other way around

Colour scale is showing velocity. I'm using Paraview with the surfaceLIC plugin. The direction of the traces can be misleading without also seeing the geometry lying close to the slice plane, especially in this case, as there are strakes near the slice plane in all 3 examples.

[variante]

So, CAEdevice, your diffuser produces about 3000N of downforce. It seems to have by far the widest low pressure peak of the mall, but it might be because the image was taken very near to a strake?

Further considerations: my diff semms to exctract a little more air from the floor. JJR's diff has a drastic transition but doesn't stall: time for me to be more aggressive with that transition

[CAEdevice]

I don't know the downforce generated into the diffuser, I only measure the df of the flat part of the floor (about 2400N).

Yes, the extension of the low pressure area is influenced by the strakes, very near to the section plane.

[machin]

CAEdevice:
http://www.khamsinvirtualracecarchallen ... er_rd1.jpg

JJR Racing:
http://www.khamsinvirtualracecarchallen ... er_rd1.jpg

Variante:
http://www.khamsinvirtualracecarchallen ... er_rd1.jpg

Who can explain why those diffuser shapes are best then? Is it s a function of the rules, i.e. "Diffuser can't start until X=??mm and Diffuser surface cannot be more than Y=??mm above the flat floor"; or would you still use that diffuser shape even if you didn't have those rule restrictions?

EDIT: Without knowing the actual answer I would hazard a guess that CAEdevice's diffuser shape gets the expansion as soon as possible (to move the COP forward) whilst still being a smooth transition and obeying the max height rule at the tail end of the diffuser...??? Or am I way off here?!

[CAEdevice]

Hi Machin.

To completely understand how my diffuser work it would be useful an horizontal section too. I can say that the low pressure area is not much extended if viewed from above.

About the design limits: the presence of the rear suspension is the greatest issue.

About the performance: the general section on a plane parallel to simmetry plane influences the performance less than inwash effetcs.

[RicME85]

FYI machin, this is what we have to play with:


Green area - completely flat section of floor.
Blue area - any visible bodywork from below can only be a max height of 100mm
Orange area - any visible bodywork from below can only be a max height of 320mm

The section at the back and far right show the space you can create the diffuser, albeit the top line is the maximum height so the underside of the diffuser would be a bit lower to allow for minimum thickness.
As you can see the suspension template dictates the early shape of the diffuser which is why these guys have gone for the hybrid solution

[variante]

would you still use that diffuser shape even if you didn't have those rule restrictions?

Yes, we would. They would only be a bit more aggressive.
The philosophy behind that shape is this:
- First Stage is concave to extract tha largest amount of air from the floor, thus decreasing the pressure where the bodywork is nearest to the ground.
- Second Stage is convex to keep the the pressure under the diffuser decently low, but also to virtually continue the "expansion" behind bodywork's physical extention (an expansion coupled with gurney flaps and RW upwash)

Note that:
- when i say expansion i do not actually mean expansion as those are not nozzles...
- concave and convex shapes work the same way, however concave allows for a more "violent" decompression; its AoA is limited either by stall or by suspensions templates...
- the reason why conventional wings aren't shaped like a concave diffuser is because they have upper surfaces...
- placing the low pressure peak near the flat floor (as our diffusers do) decreases drag
- since the lenght (=area) of those diffusers is quite great, it would be a good idea to take advantage of it: that's why there's the convex stage too. The greatest performance is achieved when the perfect compromise is found.
- the concave stage allows for a far better vortex formation

I would hazard a guess that CAEdevice's diffuser shape gets the expansion as soon as possible

The other way around. Concave has its low pressure peak placed a lot forward, thus it can interact (=ectract) more from the floor. The convex (more or less Matteo's one) distributes its low pressure along its lenght.

[CAEdevice]

We also had to manage the presence of the wide wheels. Differently from F1 rules, the diffuser is larger than the inner distance between the wheels. My diffuser, if seen from above, is far from being straight. So "concave" or "convex" definitions are not so strictly related to the progression of the volume expansion.