[MVRC] Mantium Virtual Racecar Challenge 2018

[BlakjeKaas]

Congrats to Variante for the win!

It is interesting to see how the times at the top are so close, while the cars differ so much in drag and downforce.

My submission had a slight issue with the engine intake, so I could at least go an easy second faster, if I can use full engine power. At least cooling doesn't seem to be an issue anymore

I'll hope to get a <80s car for Sao Paolo.

The results are a bit worse than I expected. So, let's wait for the report to see what is going wrong.

My results seem to be off as well by about 5%, but this can be easily explained by running the simulations on the fast setting, so I'm not that surprised.. (both C_D and C_L were higher by 5%, making the end result about the same)

[machin]

It is interesting to see how the times at the top are so close, while the cars differ so much in drag and downforce.

There was a Monaco F1 race a few years back where one of the drivers smashed his front wing off meaning he effectively had almost zero downforce (all that rear downforce is worth nothing if your front wheels won’t turn in)... he was running only 3 seconds per lap off the cars in front...

Is 3 seconds a lot or a little? It doesn’t sound like much, but across the start finish line it is going to be somewhere around 200 metres lost ground per lap...

[BlakjeKaas]

It is interesting to see how the times at the top are so close, while the cars differ so much in drag and downforce.

There was a Monaco F1 race a few years back where one of the drivers smashed his front wing off meaning he effectively had almost zero downforce (all that rear downforce is worth nothing if your front wheels won’t turn in)... he was running only 3 seconds per lap off the cars in front...

Is 3 seconds a lot or a little? It doesn’t sound like much, but across the start finish line it is going to be somewhere around 200 metres lost ground per lap...

Well, 3 seconds is rather a "lightyear" in racing, but since it's quite hard to overtake on Monaco, a 3 second deficit doesn't have to mean a lot of trouble if you're already in front.
I suppose you mean the 1997 Monaco race where Mika Salo finished 5th without doing any pitstops and with a broken front wing? His 3 second deficit could partly be explained by him driving in a Tyrell, which was already ~1,5 seconds slower than the top cars anyway. Furthermore, since he was driving the whole race on a single set, I'd suppose the tyre wear didn't help lap times as well..
A more recent example is Daniel Ricciardo, he lost his complete hybrid system this year, but went on to finish the race in 1st position because the other cars couldn't find a spot to overtake him even though the race pace would've been ~3-4 seconds faster.

However, my point was actually that the designs of the top 3 cars differ a lot in drag & lift coefficient, but the results however are very close.

It's also interesting to note that TalnoRacing designed a car that only had a L/D of ~1.5, but due to the low drag coefficient was able to finish relatively high. The question remains whether a low drag design philosophy will be the way to go for the upcoming tracks or if a high downforce approach will work better.

If my eyes don't deceive me, the next race will be in 1,5 months, giving me a lot of time to try out different concepts, and I might even be able to focus on a certain design philosophy.

[variante]

It is interesting to see how the times at the top are so close, while the cars differ so much in drag and downforce.

Well, if you’re bad at something, you must compensate with something else.
Still, the fastest possible laptime can be achieved with one and only one configuration. I don’t think that for Sepang it is mine, but I’m sure Machin and his software know which one it is

[machin]

It's not really possible for me to define an "optimum" configuration because there isn't one... The graph below might help somewhat:



Basically each track has a characteristic set of lap-time iso-curves, (shown in light grey), the gradient of which depends on the track layout. Regardless of the track the lower your drag and the higher your downforce the faster your car will go (i.e. the further down into the bottom right corner of the graph you are the better your lap time).

Now what you could do is plot on that graph some data points representing different versions of your car. I've named them "config.1, 2, 3" on the graph. You are looking for the configuration that is closest to the lowest lap-time iso-curve. In the example above config.2 and config.3 look like they will yield very similar lap times... (Rather than drawing the graph, just put your car data into Virtual Stopwatch and see which one is quicker). You would then pick the fastest configuration and then you would need to do the hard CFD work developing it to move it further down towards the next lap-time iso-bar,i.e. lower drag and/or increase downforce.

In the lower output box on Virtual Stopwatch you will see a set of data which tells you whether your current configuration would be faster with 20% increases or decreases in drag and downforce... this can help you decide which way to develop your car... although as Variante said in a previous post: with your own specific car design you might find it easier to do one or the other... it really is specific to your particular car... but a good indicator of what is possible is to look at your competitors:

Do they have a lot more downforce than you?

or

Do they have a lot less drag than you?

This might help you decide which way you need to develop your car...

www.competition-car-engineering.com/MVRC_Timing.htm

[etsmc]

any word on official numbers for each team for the last race?

[TalnoRacing]

It's also interesting to note that TalnoRacing designed a car that only had a L/D of ~1.5, but due to the low drag coefficient was able to finish relatively high. The question remains whether a low drag design philosophy will be the way to go for the upcoming tracks or if a high downforce approach will work better.

My intention was a low drag design, but I also intended to have decent downforce, definitely more than I ended up with.

I think I do not have a basic understanding of how my car works. Or at least, it works in ways I did not expect.

If I understand the below picture correctly, I am sealing the front of the floor and cutting off supply of air to the diffuser.


This results in no diffuser activity:


There is stagnation of air behind the front fender which in turn seals the side of the floor, preventing air from flowing into the diffuser from the sides.


I don't believe my concept is flawed, but it does require A LOT of refinement.

[machin]

Variante or Matteo will be along shortly I’m sure.... my guess; your rear diffuser looks quite steep, and has no strakes to invigorate the flow and keep it attached to the roof of the diffuser... that’s where I would start...

Similarly your front diffuser looks like it is stalling too...again, strakes (vortex generators) or a high pressure bleed to keep flow attached to the roof...

[CAEdevice]

Both the diffusers are stalling, and (machin is right) rear diffuser angle of attack is too steep. 20-25° are ok if you can feed it with enough air, your angle looks bigger than 30° and the flow looks not enough.

Finally: you need at least two strakes to control vortices into the diffuser and partially seal it.

I think I could help more with a section showing velocity instead of pressure.

[LVDH]

Ok,
I finally found the time to send out the reports and add the results to the challenge page. While doing so, I noticed a bad mistake with the Talno entry. I copy pasted the wrong cooling flow number to Richard. His car performed much worse than previously posted. Sorry for this.

As for the tech review of the cars, it might be worth contacting some veterans of this forum, like "Vanja #66" (who already wrote articles for f1 technical) or "jjn9128".
May I suggest a partnership with F1 Technical? Such review could be published on its main page as it already happens with other kinds of tech articles, with great benefits in terms of visibility.

It would be great if you or someone else could contact these people. I am a bit busy and would like to spend some time on improving the software. Let me know if one of you can take over this small but important task.

[CAEdevice]

Thanks LDH, now there is lot of data on the official website.
I noticed that my car too was a bit less efficient than expected (6.07 vs 6.15 not a big difference actually).

[Ft5fTL]

That 1.07 sec penalty...

Reminds me of this:

[variante]

Yeah, I immediately thought that penalty was completely wrong too. Obviously, I didn't want to complain about it in your place.

[variante]

My intention was a low drag design, but I also intended to have decent downforce, definitely more than I ended up with.

You should play safe with the most delicate parts, such as the first half of the diffuser, and be more aggressive with the others.



BTW nobody forces you to use a concave diffuser. A less intimidating convex or straight diffuser will do quite a good job too. Keep it simple until you can understand it and control it. The low pressure on the diffuser's throath should be vastly superior to what you've got now. Finally, a good diffuser gives the best when the rest of the car is helping it.

[machin]

That 1.07 sec penalty...

Reminds me of this:
https://www.youtube.com/watch?v=PK4IxI_nfaI

Yes; the penalty itself exceeds any gain you got; it’s not there to compensate, but to deter future transgressions... checking you’re inside the bodywork volumes is an easy job, so the penalty could be easily avoided. Also worth noting that this car also failed the volume check (in another area) in round 1, and was let off with a warning that time...

As it happens it had no effect on the results... this time... but you should all take note!