[MVRC] Mantium Virtual Racecar Challenge 2016

[LVDH]

Hi, very good visualization.
You almost got it right. In MantiumWFlow a monitoring_surface and a special_bc are two slightly different things.
So just not to confuse anybody here. What Matteo is showing are the engine inlet and exhaust outlet surfaces. Both go into the special_bc folder. And how he is showing it is exactly how it should be done. The engine inlet surface is ahead of any other geometry and corresponds to the inlet scoop. He did the same on the exhaust. By constructing the geometry this way the vehicle body is closed and there are no openings at the engine inlet or exhaust outlet even if you would somehow forget to add these two special surfaces. By keeping a distance of about 50mm there is no chance that cells of the volume mesh get snapped to the wrong surface.

[CAEdevice]

Thanks.

I add a question to the list (sorry to be persistent, but the first race is near). How would be measured the flow into the heat exchanger? Will be considered the "inlet" or the "outlet" surface (in my case they almost converge, but they are not showing exactly the same flow)? Maybe using a single surface in the middle of the heat exchanger would be simpler (if it is possibile to put a monitoring surface there).

[LVDH]

That is almost how it will be done. I have been running tests with a surface in the middle of the heat exchanger. Also I have been running tests on different triangulations of these measurement surfaces. They showed that as soon as you have a uniform mesh with less than 10mm triangles all three surfaces show the almost same result.
If you have less resolved surfaces you will notice that the inlet and outlet show significantly different results. Most of you seem to have a recirculation zone in front of the heat exchanger. The heat exchanger itself straightens the flow very much. This makes it easier for the interpolation algorithm to get the values on the outlet surface compared to the inlet. But as my tests show this is only true for poorly resolved surfaces as I already said. As soon as you take care of a good triangulation everything is fine.

[CAEdevice]

Ok, now I apply a super fine accuracy to all monitoring and bc surfaces. The simulation is ok... the big problem is now the car: the cooling ducts are not working well, I can't obtain more than (0.88x2) 1.76m3...

[cdsavage]

Hi, very good visualization.
You almost got it right. In MantiumWFlow a monitoring_surface and a special_bc are two slightly different things.
So just not to confuse anybody here. What Matteo is showing are the engine inlet and exhaust outlet surfaces. Both go into the special_bc folder. And how he is showing it is exactly how it should be done. The engine inlet surface is ahead of any other geometry and corresponds to the inlet scoop. He did the same on the exhaust. By constructing the geometry this way the vehicle body is closed and there are no openings at the engine inlet or exhaust outlet even if you would somehow forget to add these two special surfaces. By keeping a distance of about 50mm there is no chance that cells of the volume mesh get snapped to the wrong surface.

On this point, the image shown at the bottom of page 7 in the rulebook is confusing - the model there appears to show a cooling inlet, but this is how the engine inlet and engine exhaust should be submitted. Obviously for the cooling inlet and exhaust (the parts in the "monitoring_surface" directory), that image does not apply, there should be an unobstructed duct leading to/from the heat exchanger, with the monitoring surfaces placed so that they fit within the duct at each end.

[LVDH]

Ok, now I apply a super fine accuracy to all monitoring and bc surfaces. The simulation is ok... the big problem is now the car: the cooling ducts are not working well, I can't obtain more than (0.88x2) 1.76m3...

I have to admit getting the cooling to properly work on the MantiumRacer which I use for tests is not easy either.
I found some information on this www thing when searching for "radiator duct design", however nothing was so well explained that it was worth posting here and with the next race happening in less than a week I do not want to give any advice on this (ok, other than the internet search I just recommended).
Now that I do not have to go crazy about design details anymore I am really looking forward to receiving your designs and see what you guys come up with. And if anyone else is struggling, just remember it is the same for the others as well.

[LVDH]

Hi, very good visualization.
You almost got it right. In MantiumWFlow a monitoring_surface and a special_bc are two slightly different things.
So just not to confuse anybody here. What Matteo is showing are the engine inlet and exhaust outlet surfaces. Both go into the special_bc folder. And how he is showing it is exactly how it should be done. The engine inlet surface is ahead of any other geometry and corresponds to the inlet scoop. He did the same on the exhaust. By constructing the geometry this way the vehicle body is closed and there are no openings at the engine inlet or exhaust outlet even if you would somehow forget to add these two special surfaces. By keeping a distance of about 50mm there is no chance that cells of the volume mesh get snapped to the wrong surface.

On this point, the image shown at the bottom of page 7 in the rulebook is confusing - the model there appears to show a cooling inlet, but this is how the engine inlet and engine exhaust should be submitted. Obviously for the cooling inlet and exhaust (the parts in the "monitoring_surface" directory), that image does not apply, there should be an unobstructed duct leading to/from the heat exchanger, with the monitoring surfaces placed so that they fit within the duct at each end.

Indeed a bit confusing. Matteo's images are a very good reference though. Just do not get confused by his names. The engine_inlet and exhaust_outlet have to go into the special_bc folder and not to the monitoring_surfaces.

[CAEdevice]

The problem is not to obtain a good cooling flow: (2x2) 4m3/s has been the first result. The challenge is to have a good balance with a good cooling flow: I have the inpression that the rear diffuser is too big and the rear df is difficult to be balanced by the front wing/diffuser.

I am struggling with the front wing up-wash... I can't wait to see the cars of the other teams. Monaco will be a test race for me: I think I will have to design a new car to replace the one I designed for the KVRC opt1, but it will be ready after the first race.

About the names I wrote on the picture above: am I wrong or the engine intake/exhaust surfaces are both for bc and measuring of the pressure integral? I can change them in the next days (tomorrow I will be oit of office).

[LVDH]

About the names I wrote on the picture above: am I wrong or the engine intake/exhaust surfaces are both for bc and measuring of the pressure integral? I can change them in the next days (tomorrow I will be oit of office).

You are right, they do measure these pressure integrals. But mainly they are a boundary condition for the CFD simulation. The engine intake sucks air and the exhaust outlet blows air into the domain.
Measurement_surfaces are completely passive, they only measure.

[etsmc]

Well with the first race only days away i have realised i have gone on holiday without the files for my car from KVRC
Have a laptop with me so decided to download the rules and parts file and set about designing something just as a concept, don't have high hopes for it but better to enter something rather than nothing.
Its a different concept than the car from KVRC but no doubt round 2 will probably see the KVRC car return.
But here is the MVRC_MM01

No CFD as i cant access my work machine to run the test, fingers crossed.

[LVDH]

Hi,
cool car. I always like new concepts.
You have made another big mistake. I just checked everything. Right now it seems as if you did not sign up yet. Please do this as soon as possible as I want to send out an email to all participants tomorrow.
Everyone else who wants to race, should check if his team is listed:
http://mantiumchallenge.com/mvrc-the-teams/
If you do not find your team please sign up again, maybe I lost you somewhere.

[CAEdevice]

Hi, I have "frozen" the development of my car, beacuse I will be too busy in the next days.

My advice is to accurately define the algorithm used to compite the engine power and to publish it "officially". I have the impression that it will be the most important factor to determine the winner

PS: At the moment I am not sure to reach 50% of the required flow...

[CAEdevice]

Well with the first race only days away i have realised i have gone on holiday without the files for my car from KVRC
Have a laptop with me so decided to download the rules and parts file and set about designing something just as a concept, don't have high hopes for it but better to enter something rather than nothing.
Its a different concept than the car from KVRC but no doubt round 2 will probably see the KVRC car return.
But here is the MVRC_MM01
https://mercurymotorsport.files.wordpre ... c_mm01.jpg
No CFD as i cant access my work machine to run the test, fingers crossed.

The big inlet is very very interesting, but I am too late to copy it

The small additional front wing will not help very much...

[LVDH]

etsmc is MercuryMotorsport? If yes, then you are obviously signed up, sorry for any confusion.

My advice is to accurately define the algorithm used to compite the engine power and to publish it "officially". I have the impression that it will be the most important factor to determine the winner

Good advice, from all the big and small things I had to get done, this is one I was not able to do so far. I am sorry about this. After simulating all of the cars I will talk to Chris and Richard and find a good solution. At least the advantage of this will be we will immediately see what performance you guys can achieve. And we will make sure to find a good formula that does not mix up the ranking in any crazy way. During the KVRC time a lot was done based on good common sense and I hope it can stay that way.

What is good is that I have finally been able to solve the car upload issue. When logging into the page you will find a new entry on your personal dash board. If not you might have to log out once first. I will send an email out to everyone tomorrow.

[CAEdevice]

Thanks! I think that this detail is very important (last year there was a part of the engine power not affected by any reduction, maybe 50%, and the other part reduced by a function that, in may opinion could have a linear dependence on the effective cooling flow).

By the way: indipendetely from the weak points of my design (but I will have a new car for the second race) the simulation of the cooling properties makes the challenge much more interesting. What about simulating the brake cooling ducts next year? They could be included into the provided parts (wheels hubs) and the teams should design the car body in order to feed them (I think it would be easy to consider the brake power in VistualStopWatch).