[KVRC] Khamsin Virtual Racecar Challenge 2016

[graham.reeds]

I like the idea of an lmp2 series for those of us who want to compete but not particularly adept at using CAD.

A tutorial using a freely available package would go a long way to help out those starting off.

[CAEdevice]

Here is a link to download a free and "serious" CAD software.

http://it.ptc.com/product/creo/elements ... ng/express

It is a "modern" edition of the old One Space Design. I prefer other (not free) tools, but I think that it is a big improvement if compared to SketchUp. It is not easy to learn (on the contrary: it is difficult to use), but you can consider the time need to learn it an investement in new a professional skill, much more than an investement in the KVRC.

If you are a student, or a professor, my advise is to contact SolidWorks or Autodesk or PTC or SpaceClaim (or any other software house) and ask for a Student Edition.

[cdsavage]

Thanks everyone for the input. It looks like the preference is mainly to stick with something similar to 2015, with some refinements. I think this is the way to go for 2016. I'll focus on some of the specific points:


- Suspension visibility:

This has come up a few times, and it sounds like this is a bit of an annoying design constraint. It's worth considering removing these rules, but this will obviously have an impact on the appearance of the cars.

perhaps the templates can be refined to slimmer, more aerodynamic shapes which will allow smaller covers over them.

Just a minor point about this: the purpose of the bulky templates was to represent the range of motion of the suspension members. The actual parts used in the CFD simulation are much slimmer (these are available on the KVRC 'Challenge 2015' page).


- Cooling inlets:

I agree that the current approach with the large templates is limiting and we should try to make some improvements here. It's difficult to find a way around this problem though without allowing some very unrealistic designs, introducing a lot of subjectivity into the rulebook, or adding a lot of complexity. The ideal rule would be something like this:

trying to find a (simplified) relation between the pressure resultant onto the inlet/outlet cooling and engine performance

...but I'm not sure if this is the way we should go, as it only makes things even more difficult for those unable to do a lot of CFD testing. We'll also need to hear from Julien to know if this is viable from his end in terms of the additional work needed each round - dealing with the separate inlet and outlet surfaces this year has already been a big difficulty in this respect, and I think he'd prefer we go in the opposite direction, more like 2014 when we just considered inlets and outlets to be flat regions of bodywork.

You can regulate the surface area of both the first and second stage of the external template. So, as cdsavage said, the first stage may not intersect body parts or templates, but then the second stage may be a different shape, as long as the surface area remains the same, or within a certain percentage of the first stage.

Could you elaborate on this? If I'm understanding it correctly, this would be pretty easy to cheat by increasing the surface area of the second template without really increasing it's size.

How would we go about judging the requirement on the second template? For example, if there was a full, conic / pyramid shaped shroud enclosing the first template and entering into the second template with only a tiny hole at the front, would this be legal, and if not, how large would the hole need to be before it would be considered legal?

The current regulations already contemplate this kind of judgment:
"There must be a realistic path for flow to reach the inlet. Obstructions or shrouds which limit the flow reaching the inlet in order to reduce drag will be judged to be illegal."
We could easily add some complementary regulations anyway.

You're right, this is a part of the 2015 rules, but I think it's a weakness which we should be aiming to move away from. The parts of the rulebook that are like this are pretty much impossible to judge in a consistent way, as there's no clear point at which legal turns to illegal. If we used this requirement inside the second outer template, then the subjective wording would become more important when judging the legality of a design.

I was also thinking about the rules and trying to make them a little simpler... so I was thinking something along the lines of:-

A. You must use the supplied chassis, suspension, wheels, engine, exhaust and gearbox parts, unmodified and in the given position.
B. The Inlet, heat exchanger, and outlet templates must remain in the given longitudinal positions and orientations, but can be a different shape as long as their area remains as given.
C. The inlet, heat exchanger and outlet templates must not intersect any of the given parts.
D. The inlet and outlet templates must not be covered with bodywork when viewed from the front or rear respectively. The heat exchanger template must be completely enclosed by bodywork.
E.The engine intake template must not be moved or changed in shape and must be completely visible from the front.
F. Bodywork must not be visible outside of the bodywork legality box.

The legality box is designed to restrict the car shape to something resembling an f1 car, but otherwise the designer is fairly free to try lots of different solutions....

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

Fixing the inlet at a certain longitudinal position would definitely simplify things. However I think specifying that the inlet must be fully visible from the front, ie an outer template of infinite length, means that the vertical/lateral position becomes quite limited. If it's placed low then it's obscured by the front wing, if it's placed outboard then it's obscured by the front wheel. If we stay with some variation on the 'template' idea then we can avoid these limitations.

This concept also does away with any rules governing the diffuser, and just limits the size by the shape of the bodywork volume, with a gap between the diffuser and the rear wing. The 2015 rule governing the diffuser geometry is complex, and it might be good to be able to get rid of this. For this to work though, the vertical position of the rear wing must be more heavily restricted. What does everyone think about making this tradeoff, ie the diffuser geometry is completely free, but there is an exclusion zone at the rear of the car between about Z=350 and Z=750, with low-mounted rear wings not allowed? The only big problem I see is in how to deal with the bodywork behind the rear wheels.


- Supplied parts and a secondary class

There is actually a set of parts supplied this year, this is available on the 2015 Challenge page on the KVRC site. This includes a nose, tub, engine/gearbox cover (ie the entire central part of the car from front to back), as well as some fenders. I think we only had 1 competitor use these parts - maybe not everyone was aware that they were available. I avoided supplying any wing or floor shapes, but maybe we should have supplied these also. For 2016 we should definitely include the bodywork forming the inlets and outlets, and maybe some parts from a top 2015 entry as has been suggested.

I also had another idea to try and increase participation... making a "Division B" championship in which the competitors must use the normal supplied parts listed above, but also including the following:-

Nose cone
Side pods
Engine Cover
Floor

Leaving only the following parts available for user design:

Front wing
Rear Wing
Diffuser

But within the "legaility boxes" for those parts they would be free to design whatever they wanted: i.e. no limits on number of elements.

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

This should enable a more casual user to take part without them having to face the daunting task of designing the whole car.....

The intention would be that a good Division A car should be faster than the Best Division B cars, but the competitors in Division B are fighting for their own championship title... ("Division B Champion")...

I like the idea of a separate class where parts like these are mandatory, leaving just the wings and floor. If we're going to have a secondary class, I think this is the only realistic way to do it, as working to two different rulebooks is too much to manage. This also allows a competitor in the secondary class to move up to the top class if they choose to start modifying these parts.

What about the possibility to assign double (or triple) points to the last race?

I think earlier in the year you suggested a gradual increase in the points awarded each round - I probably prefer this to a big increase for the last round. However I think it's also important to think about why people enter the first few rounds but then quit - my guess is that the championship points aren't the reason, it's probably more to do with the time required to produce a competitive car. It's probably less likely that the people fighting for the top positions are the ones who stop submitting entries during the season.

Chris - how would helping with the competition work if you wanted to participate?

In the near future, when we get to the detailed writing of the 2016 rulebook, any input on that would be a big help. We also need to figure out how to approach the promotion side of things.

An idea also would be that for each round a team is selected to talk about his car, the philosophy behind it, and so on. Make a presentation of its car and his way of thinking. We need more discussions about the cars to be honest. Next season I plan on showing the guys my car in detail and giveaway its "secrets".

This leads to me saying that I wouldn't mind helping the staff and therefore putting me out of the competition if that can help attract more people. Of course I would still compete but without points.

I agree with this - this was something I was hoping we would work towards in 2015, but I've not been able to get around to it. It would be great to have someone having a look through the openfoam case files each round and creating some images to show off some of the design features of each entry and how they work aerodynamically. It probably wouldn't be appropriate for it to be done by someone participating in the championship, though.

Something I'd like some more feedback on: which rules in the 2015 rulebook are the most difficult / most time-consuming to comply with?

[RicME85]

Re promotion, surely you just leave that to those participating. They can have the choice of staying in Champ B or moving up to B. Winning B and then competing in A would be a massive jump if you haven't spent time working on ideas or just getting to grips with the software and modelling.

As for most difficult part of 2015 rules, for me its the whole cooling package restraints, mostly from the external extrusion part if the rules.

I believe time and competitiveness are definitely the key factors to lack of competitors at various points of the season. I am always surprised by the registered numbers and the actual first race numbers, there is usually a significant amount not entered.

I guess I am the one that has used the supplied parts. First race I uses all of them but with a modification to the rear quarter and then after its just been the main section as its nice and tight and we'll shaped, not really seen a need to move away from it, think there are more important areas to work on.
A floor with a diffuser would be a good addition to the starter pack, not just for an easy base to start with but also gives the less creative of us some ideas of what can be done within the regulations.

[etsmc]

With regards to the cooling inlet outer templates could you not just keep the 2 outer template rules but add a bit that outer templte 1 can be rotated in xyz by so many degrees and 2 can also be rotated in xyz the same or less. Allows flexibility on there positioning a bit but also keeps a realistic path if the rotation is only allowed to be minimal.
I like the idea of a lower grade where the car is supplied and only have to create front/rear wing and diffuser. Coyld teams then enter 2 cars, one for each championship?

[MadMatt]

Machin for me the biggest issue was cooling requirements. As I said, and as etsmc just said, I think allowing a bit more flexibility on the orientation of the faces would be a good thing. I don't mind having the flow path checked "by eye" and having as a rule to have a "realistic flow path".

[LVDH]

I also think that the biggest challenges are the suspension thing and the cooling rules.
For the suspension, I think the rule is good as it creates quite a challenge, which is good and it makes the cars more elegant. Also, from what I see, so far nobody has come up with a good solution. It seems that every cover is creating some lift. So why not give us another year to figure out something smart?
With the cooling I would also wish some more freedom. The outer templates are a bit too long now, in my opinion. Also it would be nice to be able to rotate the cooling exit. This way we could come up with more elegant solutions. Maybe inlet and outlet infront of the windscreen, like on some GT cars?

Also there was something else bugging me sometimes: Two cars with very different L and D but almost same lap times. Which car would you like to be driving? I think it would be the low drag one as it will be easy to overtake the high drag car. While simulating this is currently a bit out of the scope, maybe we could introduce some weight saving due to lower fuel demand based on drag?

[CAEdevice]

The outer templates are a bit too long now, in my opinion.

I completely agree, expecialy about the outer template of the cooling exhaust.
What about the possibility to use bigger outlet/inlet surfaces with shorter outer templates in addition to the present ones (or using a linear law to regulate surface vs template, considering that bigger surfaes would require a smaller dynamic pressure, at least into a reasonable range of variability) ?

Also there was something else bugging me sometimes: Two cars with very different L and D but almost same lap times. Which car would you like to be driving? I think it would be the low drag one as it will be easy to overtake the high drag car. While simulating this is currently a bit out of the scope, maybe we could introduce some weight saving due to lower fuel demand based on drag?

Last year (see 2014 thread) Machin convinced me that the fuel saving would be negligible.
Anyway: the overtaking simulation would be great, but, how to consider it? Maybe using a sort of AI or trasferring our car data into a package like Rfactor2 or Assetto Corsa?

[MadMatt]

Also there was something else bugging me sometimes: Two cars with very different L and D but almost same lap times. Which car would you like to be driving? I think it would be the low drag one as it will be easy to overtake the high drag car. While simulating this is currently a bit out of the scope, maybe we could introduce some weight saving due to lower fuel demand based on drag?

There is not 1 rule with high DF being better than low DF or the other way around. Check Monza for instance. One year Vettel had a high downforce car and yet he couldn't be catched, so low drag doesn't always mean overtaking chances (especially if you cannot get close enough on the corner before the straight). Plus having more downforce makes things easier for the driver (your car is not sliding all the time).

[LVDH]

Wouldn’t the car have greater tire degradation due to higher loads with higher downforce?

[MadMatt]

Wouldn’t the car have greater tire degradation due to higher loads with higher downforce?

For sure but unless the track is an oval, you will be faster around the whole lap than your opponents. Up to you to see if 1) you need to do 1 more pitstop (it could be that the tyres are hard enough to cope with extra load), 2) you are fast enough to remain in the lead after that pitstop.

In these simulations the issue for us is to know the trade-off between having maximum downforce but not very efficient, or slightly lower downforce but much more efficient, because the higher the downforce, the more difficult it is to make it efficient.

[LVDH]

I know, I know.
I guess my last suggestion would make things too complicated. I am surprised that taking the fuel savings into account would be neglectable. But if that is the case why add more complexity?

[machin]

The fuel saving thing; imagine you have one car exiting the corner at 50mph and accelerating down the straight... It might reach 200mph and cover the straight in 5 seconds... Another car might exit the corner at 45mph, but have lower drag and reach 205mph... Its terminal speed is higher, but it started slower so it might take a similar time to cover the straight... if both cars use full throttle down the straight, and for the same duration, they'll consume the same fuel (regardless of what the drag of their car is)... So low drag doesn't translate to a hugely lower fuel consumption on race cars; lower drag just means higher acceleration and possibly higher speed for the same fuel consumption... The main saving comes in the corners where the car cannot use its "slipperiness" to go faster, since it is grip limited... But all cars are at part throttle in the corners and therefore fuel consumption is relatively low compared to the straights anyway...

With road cars, where laws force you to drive at prescribed speed limits at all times, fuel consumption does significantly decrease with drag simply because the driver cannot accelerate or go faster.. Essentially a lower drag car just needs less throttle (hence fuel consumption) to travel at that speed.

That all said, it is relatively easy for me to calculate the fuel consumptions...

...I'm happy to make the change to a fuel allocation per lap rather than a simple power curve if people want? I.e cars using less fuel will be allowed a power increase to use up the remaining fuel... The impact on you guys is simply that the "lines of equal laptime" on the chart I posted in the 2015 thread will be slightly lower gradient (i.e promoting a lower drag design). But there would still be a low downforce configuration and a high downforce configuration which results in the same lap time...

[machin]

Similarly for the tyres; I could introduce a wear rate such that as the tyres get used the coefficient of friction reduces; more load on the tyres means higher wear rate, so by the end of the lap the higher downforce cars see a bigger reduction in tyre coefficient...?

Again, the impact for you guys is that the gradient of the "lines of equal laptime" on the chart in the 2015 thread reduce slightly so that high downforce configurations aren't quite as good as they are at present, but again there would still be a high downforce configuration which achieves the same lap time as a low drag configuration...

I.e. We'd be moving the goalposts slightly, but not significantly changing things from the competitors point of view.

[RicME85]

Should we not work on the the CFD solver etc before we start thinking about these sorts of things?