Dynamic Wind Resistance

This thread is about wind resistance on cars.

I think it would be a good idea if wind resistance was effecting cars as they drive. If a door opens, bonnet or boot then depending on how fast the car is going it will be blown back shut. The speed will be effected and the car will slow down if the bonnet is up and eventually will snap the bonnet back shattering the windscreen. A convertible roof will be broken and pull back if it is transforming while the car is in movement.

Thanks for reading, let me know what you think about this idea.

 

I definitely like this idea, especially for this:

This obviously wouldn't apply to a slower car, but would be awesome for fast cars. Also, I like the idea of damaged panels being blown off of the car (like an open hood or bent quarter panel).

 

The game RoR already had this... basically each node itself would be subjected to wind resistance from the vehicle moving through the air, this means that it automatically simulates wind resistance for doors being open and what not... however, most people use fusedrag in RoR so the effect on the interior nodes doesn't appear as large allowing vehicles of lower weight and power to hit higher speeds...

Seeing as Beaming is based on the same principles as RoR I am guessing it will work pretty much the same.

 

I've never seen it in RoR. I was driving and my bonnet lock was broken and it didn't come up. My doors didn't shut either.

 

It depends on a couple of factors in RoR that real life doesn't have to deal with...

First off it depends on how many nodes are in the surface that is experiencing the wind resistance... obviously the more nodes, the more resistance.

It depends on weather the vehicles are weighted correctly... for instance, the Dodge Ram that was released a few years back weight something like 11 tons... which isn't right, so to get the hood to fly up under that weight you would probably have to go over 250mph to get enough resistance.

It depends on weather the vehicle is using Fusedrag as I mentioned earlier. When using this, the nodes inside the defined radius are basically void of all wind resistance to simulate the nodes being inside the vehicle. This enables vehicles to have realistic drag coefficients and be able to hit realistic speeds without having ridiculous amounts of power and weight to push through the extra wind resistance.

If the first 2 factors are combined correctly, then the vehicle will exhibit realistic wind resistance, however, usually they are not... I am guessing that you are talking about one of the Gavrils however... which they do have the first two correct...

Unfortunately with light vehicles it is almost impossible to make one handle correctly without Fusedrag. This is why the Gavrils may not exhibit the correct behavior, but there is basically no other way. This means that the nodes basically have almost no wind resistance. So to add realism you could add wings to each one of your fenders to simulate the surface through which the air is hitting.

This to me means that the Wind resistance in RoR is over calculated on each node, or there should be a better way of defining the fusedrag. I hope that this calculation g for BeamNG will be tweaked so we don't have to do little shortcuts like fusedrag to get vehicles to work correctly.

 

You definitely seem to know more it than me, but I'm still not convinced that RoR had any wind resistance. In every car I have, wind resistance seems nonexistent The hood will sometimes fly open when I hit a bump or fall, but it never flies up due to wind. For example, if I'm driving at 100mph. and I open the hood, it will stay open at a set degree (rather than instantly being pulled toward the windshield). You can even close the hood without any problems. The same happens with doors. If I drive at 100mph. with my doors open, they don't close. If there was wind resistance, the doors would be slammed shut. Although your description sounded intelligent, I'm not sure it's correct. I personally saw no sign of wind resistance in RoR. This is definitely something BeamNG should implement into the final game.

 

Oh, it is applied due to the shape? I guess I just naturally assumed that it was applied using all the nodes because... I don't know why I just assumed I guess... interesting to note though

 

Is there any published paper about the simulation method you are using? It must be interesting. I've been thinking about such thing before.

 

Apologies for commandeering and for bumping this 2 year old thread. Also please excuse my ignorance regarding technical details.

I have searched and come across several threads like this that mention aerodynamics & wind resistance, but only a couple that talk specifically about downforce in Beamng, and what I've seen indicates that there isn't any (at least not without specialty surfaces). I've also done some of my own testing with the Bolide that seems to confirm that, but that strikes me as a little odd considering you do have all these other elements of wind fluid dynamics. Am I missing something, or is downforce something that is separate/more complicated than I'm imagining? On that note actually, how about drafting (driving behind another car to fall into the slipstream & be shielded from wind resistance)? Just curious about whether these will be implemented at some point. Thanks,

 

Ok, first off I would like to congratulate you on being the first person that I have ever seen that has under 50 posts to actually use the search bar and not just make another pointless thread. You only have two posts and I think you should be a senior member already, so bravo!

Secondly, Although people complain about bumping threads all the time, it usually has to do with a mod that someone comes along and says "CAN I HAZ BETAZ!!!1!11!" (slight over exaggeration) even though the thread hasn't been posted in for two years and is completely dead. This time however you actually added to the thread constructively, so I think everything will be fine. Also, congratulations for actually being the first person to actually notice they were bumping when they were bumping... again a rare sight!

Anyways, as far as I know (here comes my RoR days again) Drag and Down force are calculated completely differently in BeamNG (and RoR). As explained by Tdev above, the wind resistance is calculated by the shape of the collision shapes on the vehicles. All of that basically (assuming again) goes into calculating just the drag of the vehicle through the atmosphere.

As for down force however, (in the olden days of RoR) down force was calculated by adding a physical wing to the vehicle. You literally picked 4 (technically 8) nodes for which you wanted for form a wing out of and that would then produce down force (or lift if you were making a plane or had no idea what you were doing).

Now if this is not the case in BeamNG (which it really shouldn't be, it should be calculated along side the drag) then I am sure someone will come along and correct me, but if it is, then the reason that the cars seem like there is no down force is because they were made without any wings probably. I haven't dove into the files myself, nor do I have a good enough understanding of JBeam to even know if I am right, but that's my two cents.

Edit: Just noticed that you mentioned drafting. To my knowlege this is not calculated at all in RoR or BeamNG mostly because of the fact that you only every usually have one car driving so why bother calculating drafting. In the future I am sure it will be added once we get a real racing and traffic AI system and have online game play, but for the moment I am pretty sure it is not in the code. If I am wrong again, someone please correct me on that.

 

It seems to work on a similar principle. I've got no idea on JBEAM, but wings added in by nodes and only nodes (no mesh or coltris if you usually need those) seem to produce their accompanying force. The pidgin produces enough lift to fly at about 170Kmph I think.

 

When? Going at 70mph with a car bonnet curled up facing the airstream and the latch broken and it'll happily remain closed. None of the default cars seem to anyway.

 

Wind resistance is determined by the collision triangles, and like a polygon, the collision triangle has a direction, or normal. If the triangle is facing the direction of travel, the wind resistance will be maximum. Conversely, if it's facing opposite the direction of travel, wind resistance will be zero. Since the coltris on the bonnet are facing outward, a bonnet curled up will expose the inside faces of the coltris, effectively negating all aero forces.

 

Thanks sappy for the correction. I thought it was still based off nodes. Backwards coltris are also the cause of some of the inter and intra vehicle collision issues. They seem to be buch better in the early build though.

 

Downforce or lift is just the drag force applied in a different direction if the triangle is at an angle. Triangle side doesn't matter, it works the same both ways.

Wings and other aero devices tend to be more efficient than an angled flat plate, so they should have the drag coefficient set over 100. Unfortunately that gives them more drag than they should have. A whole car has less drag than its parts individually, so the parts need lower drag coefficient as a compromise. Otherwise the car has too much drag and lower top speed. That could be why individual parts don't work as expected in wind. Doors, hoods and other surfaces that are parallel to wind when driving straight can have closer to 100 drag coef.

 

so that explains why individual car parts have such low air resistance. I'd need to be going 1000mph for the hood to fly up as i discovered when I modified the moonhawk's engine to have a negative drag coefficient. By the way guys, negative drag coefficients are hilarious.

 

Try negative friction on grand theft auto, utter hilarity.

 

Does that mean I don't have to waste time making sure all my coltri nodes are defined CCW?

 

I think miura meant it doesnt matter for drag. It does for collision, back of a coltri is non solid, I think they should be double sided personally.