Khamsin Virtual Racecar Challenge 2015

Post here information about your own engineering projects, including but not limited to building your own car or designing a virtual car through CAD.
User avatar
machin
162
Joined: 25 Nov 2008, 14:45

Re: Khamsin Virtual Racecar Challenge 2015

Post

In theory, more realism sounds good, and it would be relatively easy (once we have the aero data) to take cooling flows into account on power (My final year Uni project was working with Ilmor to optimise heat exchanger and cooling duct design for their petrol engine powered Voodoo target drone, so I have some data to work with). You could also account for different suspension geometries relatively "easy" too...

...The downside is that to be competitive you'd need a team of people working on different aspects of the car... Running lots of iterations... And I think people are already struggling to find the time to do all the basic aero stuff...???

Personally I don't think we lose that much by being highly prescriptive on some aspects of the design (like suspension engine air intakes, etc); in real life as engineers we're often given constraints when we're designing things due to costs or availability ("You must use this gearbox with these suspension pick-up points", etc)... So having constraints to work to (like positions and sizes of inlets) could be seen as still be relatively realistic whilst ensuring the challenge remains accessible to as many people as possible...

Just my thoughts....
COMPETITION CAR ENGINEERING -Home of VIRTUAL STOPWATCH

User avatar
CAEdevice
48
Joined: 09 Jan 2014, 15:33
Location: Erba, Italy

Re: Khamsin Virtual Racecar Challenge 2015

Post

is there a way to measure a pressure or a flow through a surface with openfoam+paraview?

User avatar
LVDH
46
Joined: 31 Mar 2015, 14:23

Re: Khamsin Virtual Racecar Challenge 2015

Post

CAEdevice wrote:is there a way to measure a pressure or a flow through a surface with openfoam+paraview?

Yes, no problem at all, even just using OF.

The main problem is that some people will struggle with submitting valid CAD data.

Then there is a certain issue with adjusting engine power to cooling air flow. I think the issue is not as big as it might sound on first sight but people will complain. On the other hand we are looking at sections forward and behind the heat exchanger which are staying completely straight. Both options are not perfect. I think for the competition it is working very well so far. While I do not like these open body designs too much I think its fine. If KVRC wants to progress forwards, then I think finding a way to have aero also affect other aspects of a car, as it does in reality, then that would be the way to go. But that is a decision for the people organizing it. The team is doing a very good job, and having to get people to create valid CAD data for OF will be a pain in the certain hole they have to be willing to take. And most non-OF people will have problems understanding what OF wants.

User avatar
CAEdevice
48
Joined: 09 Jan 2014, 15:33
Location: Erba, Italy

Re: Khamsin Virtual Racecar Challenge 2015

Post

It's a very interesting project for KVRC 2016, but at the moment I think that the priority should have a reliable solver (framework) also for local run.

In the future it would be interesting to look for a partnership with CAD software houses (in the past I took part to contest where a limited edition of the CAD software was provided by the sponsor, Spaceclaim).

I was wondering wich CAD tools are we using. CAEdevices uses SolidWorks for general modeling (or Creo) and a very old release of Spaceclaim to manage stl (skp in kvrc 2014) conversion and quality.
I'm still trying to use OCCFD (official framework) but with some meshing issues.

julien.decharentenay
julien.decharentenay
10
Joined: 02 Jun 2012, 12:31

Re: Khamsin Virtual Racecar Challenge 2015

Post

Hi, Sorry I have been out of the loop. I have been property hunting in London. Quite stressful and time consuming. Hopefully I am in the process of securing something - the first to get the family over.

Calculating the flow through the cooling and engine ducts would be great, and quite a challenge. But, as Matteo mentioned, getting a robust solver is a first step. I have a feeling that the free openFoam distribution for Windows lack some stability. Would anyone be game to build a new version based on the instructions at http://www.symscape.com/openfoam-2-2-x-on-windows and make it available?

User avatar
CAEdevice
48
Joined: 09 Jan 2014, 15:33
Location: Erba, Italy

Re: Khamsin Virtual Racecar Challenge 2015

Post

Why don't you ask directly to Richard Smith from Symscape? They could became a sponsor just providing the engine (Windows compiled Openfoam 2.4) for OCCFD.

cdsavage
cdsavage
19
Joined: 25 Apr 2010, 13:28

Re: Khamsin Virtual Racecar Challenge 2015

Post

An update on the rule change proposal I posted a while back: from the feedback, it sounds like some aspects of these rules were confusing. Here are some images:

Area template:
The K4.1 area template is a single continuous shape which is coplanar with the rear face of the inner template. The area template must cover the entirety of the rear face of the inner template, must be at least 150,000mm^2 in area, and must be no more than 600mm wide and no more than 450mm high. This template must be entirely enclosed within the bodywork and must not intersect any other parts or templates.

Image

Image
The area template is a part that you create and include in the submission (it is in the list of parts at the start of the rulebook). What does everyone think about the 150,000mm^2 dimension? I think ideally we would make this larger, but there's not been any feedback on this point yet.

There was an addition to K4.1, regulating bodywork surrounding the outer template. I'm suggesting changing this slightly:
A line drawn from any point on the the outer template upwards or downwards may not intersect bodywork in both directions. It may intersect in one direction, or neither direction. This requirement is ignored for the first 200mm of the outer template. The front suspension templates are ignored for this requirement.

Image
The 350mm rearward movement limit for the upper side impact structure could be changed to 400mm (or we could remove this template completely?), and the lower side impact structure will be removed as mentioned in the original proposal.

The other changes originally suggested:
- K4.2: Outlet must now be at least 1000mm rearwards of the inlet (+200mm). Last 200mm of outer template may intersect rear suspension templates. Inner template may intersect rear suspension templates.

- Split drivetrain template into upper/lower engine template and gearbox template. Together, these 3 parts are identical to the current drivetrain template. Rules which previously referred to 'drivetrain template' now refer to one or both of the engine templates. The gearbox template is just for bodywork to enclose.
In the interest of keeping the changes to a minimum, I think we should limit the drivetrain template split to be two pieces instead of three, by splitting it into an engine template and gearbox template. The upper and lower engine templates from the image I posted would be combined to form the engine template.

I am also suggesting that we change K4.2 so that the outer template is 900mm long rather than 1000mm, and the final 325mm of the outer template may intersect the rear suspension templates (to roughly follow the dimensions for the outer template for the cooling inlet).

Any more feedback on these suggestions? We should have a firm proposal out by around the 16th, which is when the vote will open.

User avatar
CAEdevice
48
Joined: 09 Jan 2014, 15:33
Location: Erba, Italy

Re: Khamsin Virtual Racecar Challenge 2015

Post

K4.1: 150000mm2 is ok, but also 200000mm2 could be acceptable

K4.1 "addition": it is not clear to me the reason of that rule. It would be impossibile a design like the one I used for the first race (cooling exhausts blowing under to rear suspension cover), but the car I have designed for the 2nd and 3rd race would be ok. Anyway: the wording is clear.

K4.2: ok, I would only extend the possibility to intersect the rear suspension to the last 500mm, with a total length of the outer template of 1000mm

cdsavage
cdsavage
19
Joined: 25 Apr 2010, 13:28

Re: Khamsin Virtual Racecar Challenge 2015

Post

CAEdevice wrote:K4.1 "addition": it is not clear to me the reason of that rule. It would be impossibile a design like the one I used for the first race (cooling exhausts blowing under to rear suspension cover), but the car I have designed for the 2nd and 3rd race would be ok. Anyway: the wording is clear.
This rule would only apply to the inlets. The reason for the rule: currently there are a few possibilities for reducing drag through unrealistic shrouding of the inlets. This wouldn't completely solve that issue but it would go some way towards preventing it.

User avatar
CAEdevice
48
Joined: 09 Jan 2014, 15:33
Location: Erba, Italy

Re: Khamsin Virtual Racecar Challenge 2015

Post

cdsavage wrote:
CAEdevice wrote:K4.1 "addition": it is not clear to me the reason of that rule. It would be impossibile a design like the one I used for the first race (cooling exhausts blowing under to rear suspension cover), but the car I have designed for the 2nd and 3rd race would be ok. Anyway: the wording is clear.
This rule would only apply to the inlets. The reason for the rule: currently there are a few possibilities for reducing drag through unrealistic shrouding of the inlets. This wouldn't completely solve that issue but it would go some way towards preventing it.
Sorry, I misunderstood (outlets instead of inlets).

It seems that the new rules are compatible with the old ones. I think I will anticipate their application during 4th race (no points), if approved.


User avatar
variante
135
Joined: 09 Apr 2012, 11:36
Location: Monza

Re: Khamsin Virtual Racecar Challenge 2015

Post

RicME85 wrote:I dont understand K4.1
It just says that the outer inlet template must be fully visible from the top or the bottom (or a combination of the two, like in the drawing...that's your choice).

The rule should guarantee that the inlet is fed with a realistic amount of air (supposing that an inlet fully enclosed by bodywork doesn't get enough airflow, especially if it is located, let's say, right behind the front wing).

The first 200mm of the inlet outer template is exempted from the rule (i.e., it can be fully enclosed by the top, bottom and sides)

julien.decharentenay
julien.decharentenay
10
Joined: 02 Jun 2012, 12:31

Re: Khamsin Virtual Racecar Challenge 2015

Post

CAEdevice wrote:Why don't you ask directly to Richard Smith from Symscape? They could became a sponsor just providing the engine (Windows compiled Openfoam 2.4) for OCCFD.
Good idea. I will get in touch with them.

On another note: I would keen to hear of feedback on the half-car simulation. My test are showing that it is not stable at this stage and it is unlikely that we (from the KVRC organisation side) will move over in the next few races...

Next race is only 11 days away...

User avatar
CAEdevice
48
Joined: 09 Jan 2014, 15:33
Location: Erba, Italy

Re: Khamsin Virtual Racecar Challenge 2015

Post

I tested the "half car" configuration, but the meshing issue did not allow me to verify the stability.

User avatar
CAEdevice
48
Joined: 09 Jan 2014, 15:33
Location: Erba, Italy

Re: Khamsin Virtual Racecar Challenge 2015

Post

I tested the "half car" configuration, but the meshing issue did not allow me to verify the stability.