[KVRC] Khamsin Virtual Racecar Challenge 2016

[cdsavage]

Well, with most software you could pull the mathematical surfaces to other surfaces to perfectly match.
But then for CFD you discretize them to a triangulated description. The triangles will in the end then not match anymore.
You would have to pull the surface and intersect or cut with the duct surfaces and then triangulate. But here it is already getting a bit to complicated for such an, in the end, unimportant little surface.
If we could go with your last suggestion, I think we should be good.

This is at the CFD meshing stage, so this is the same for all competitors. But if the goal is to produce a perfectly matching inlet/outlet surface prior to the meshing stage: this is trivial to do using nurbs, but it should be possible to achieve this with subdivision surfaces also if you use edge creasing. You apply a creasing weight of 1 to a loop of edges inside the duct, and then create the inlet/outlet surface from these edges. Then the outer edges of the inlet/outlet surface perfectly match the walls of the duct. The only challenge is making sure that the edge loop is planar, but you might be able to use a 'slice plane' tool to do this.

[SR71]

Now I just need to learn what this means

That means that your challenge has started. You have a lot of work ahead of you.

Legends would help us a lot to understand better what is happening. The first one, I guess, shows the pressure distribution on you car's surface. Blue should indicate low pressure. So infront of your diffuser it should be dark blue and in the duffuser light blue. Why yours is red is a good question. The design does not look terribly wrong. Maybe the simulation did not converge. You should have a look at your residual and force curves plotted against the iterations. The residuals should somewhat smoothly go below a threshold like 1e-3. Your forces should smoothly and then flat converge towards some value. If it is fluctuating like crazy you have a problem.

The histogram plot shows the chunks of forces acting on your car, here in X-direction. You should get one for drag and one for down force. In this case I am not sure which you are showing as your red diffuser is a bit confusing. In any case you will see a nice, high tower around your rear wing. It will create a lot of down force but also drag.
Maybe you could create a new thread about your car. Last year we helped, I think RicME, a lot in his own thread on his car got much better and it is fun for everyone to help and know everything better.

Definitely was running backwards, had my first glitch of the season, I'm sure it wont be my last.

Anyways, re-oriented and ran the simulation - I'll create a new thread around my car as you suggest. As this is my first season I'd appreciate any input from the community.

[CAEdevice]

@SR71: Don't worry, last year I developed my car for the 1st race without including the wheels!

[machin]

The way I use Blender, my side pod, the engine cover, the floor and the body are separate bodies.

I have a similar question I think: Do the various parts of the car need to match up perfectly, or is it OK that they intersect slightly in order to avoid gaps? Specifically in the case below I'm talking about a class B car where the front wing/suspension cover extends slightly inside the wheel fairing "supplied part". Is that allowed, or does it mess up the meshing?

[cdsavage]

The way I use Blender, my side pod, the engine cover, the floor and the body are separate bodies.

I have a similar question I think: Do the various parts of the car need to match up perfectly, or is it OK that they intersect slightly in order to avoid gaps? Specifically in the case below I'm talking about a class B car where the front wing/suspension cover extends slightly inside the wheel fairing "supplied part". Is that allowed, or does it mess up the meshing?

http://i90.photobucket.com/albums/k248/ ... zcuvgh.jpg

Doing that is perfectly acceptable - they can intersect as much as you want. The mesher generates an outer shell from the union of these parts. I would recommend trying to make each individual part a closed, solid body to eliminate any possibility of holes forming, but this isn't strictly necessary as long as all of the open edges lie inside the bodywork.

Related to this: in 2015, many of the Sketchup entries had a lot of internal faces, construction geometry, overlapping faces, etc. None of this should have affected the CFD simulation, but it makes the cleanup process for the renders very time consuming, so I would ask anyone using Sketchup to try to keep this stuff to a minimum in the submitted parts.

[machin]

Doing that is perfectly acceptable - they can intersect as much as you want.

Cool. Cheers.

[RicME85]

Are you entering a class b car or doing some investigations?
In the situation you are talking about, I use the intersect tool in Sketchup to produce a defined meeting of the two parts.

[machin]

I'm just having a play... I'm using an old version of Solidworks... it has an "extrude to surface" or "cut with surface" tool, but in a few cases I'd rather just extend into the surface and leave it at that.

[etsmc]

- 230K was the planned size for "option 2", but I agree that with the 100mm extrusion, 230K is still a bit too large. I am checking with Julien, but I would propose either:
1) 100mm extrusion -> 50mm, 230K area -> 225K
2) 100mm extrusion unchanged, 230K area -> 200K
Both with the 30 degree limit unchanged.

I would be happy with option 2.

[RicME85]

I will go with whatever as I'm not sure if I can compete straight away, if I do it will be a class b car.
Going in to hospital tomorrow for surgery Wednesday and having a second surgery some time in the next month to have my elbow replacement removed and a new one fitted (and yes its my dominant arm so using mouse and PC will be a right pain)

[SR71]

- 230K was the planned size for "option 2", but I agree that with the 100mm extrusion, 230K is still a bit too large. I am checking with Julien, but I would propose either:
1) 100mm extrusion -> 50mm, 230K area -> 225K
2) 100mm extrusion unchanged, 230K area -> 200K
Both with the 30 degree limit unchanged.

I would be happy with option 2.

I think either option works but we need more degrees of rotational freedom, I feel like all contemporary LMP1's have uniquely positioned HE's and creates greater packaging variety.

For me the perfect formula would be:

-50mm extrusion, 220K area, 45 degress of freedom XYZ

SR71

[CAEdevice]

I will go with whatever as I'm not sure if I can compete straight away, if I do it will be a class b car.
Going in to hospital tomorrow for surgery Wednesday and having a second surgery some time in the next month to have my elbow replacement removed and a new one fitted (and yes its my dominant arm so using mouse and PC will be a right pain)

Sorry about that Richard, I hope you will be back after the first race! Luckily the first one is only a test.
I if can help with CAD in anyway let me know.

[RicME85]

Thanks Matteo. I know its only a test but its good to run a car in it to get a benchmark of where you are at in the pack.
Will see what I can get done.

[CAEdevice]

Thanks Matteo. I know its only a test but its good to run a car in it to get a benchmark of where you are at in the pack.
Will see what I can get done.

If you have a STEP version of your car at the present elvel of development, and some notes about the next evolutions, I can complete it. I'm sorry I can't open/manage SketchUp geometry directly.

[RicME85]

Unfortunately I havent got anything yet. Like this time last year I am struggling to get an work started. I have opened Sketchup and made a few boxes and curves here and there but not saved anything. I think the lack of finalised rulebook is playing havok with my brain and it is thinking 'why bother' despite me knowing things wont change much and if they do it wont be anything major.
I have a concept I want to follow, annoyingly SR71 has got there first and produced 75% of the car I was hoping to produce