Educational experimenting with vehicle physics

Hello, newbie here... I would like to play around with the vehicle properties for myself, with the aim of learning the effects of the lesser-simulated properties, such as stiffness, elasticity, suspension component weight, even suspension geometry (bump steer, chamber change, roll center characteristics, etc.). I take it the beam characteristics in the jbean files allow for much of that; but one interesting point I'd like to test out, which I'm not so sure about is the elastic properties of stuff like the rubber parts in the suspension - This goes toward the question of why air cooled Porsche 911 with their ancient semi-trailing arm suspension (same as on a VW Beetle) were known to be "widow makers" prones to snap oversteer, while the same air-cooled models were actually known to be reasonably mannered and controllable in terms of it's rear suspension when in full race-car configuration (using solid uniball bushings); also, the newer 911s (ever since 993 generation) are fine too, as they received an entirely different rear suspension. I'd love to find a way to simulate this and other behavior in the game - I guess any car with semi-trailing arm rear suspensions should show similar properties.

Also, are there established ways to test such advanced parameters, i.e. simulate the resulting chassis stiffness, get information about roll centers, camber change curves etc.? Sorry if this is all very basic, I haven't found anything on the forum using engineering search terms (which I suppose would be used if someone bothered about these issues) such as "camber curve", "torsional rigidity", "bump steer"... the only tool I found in that way was nodeBeam editor, which I take is for importing models, right?

Much appreciated if anyone took the time to point me in the right direction to get started on that.

 

There is no way to get such information from the game as is, even though you could probably make a ui app to calculate all these values. But it's definitely possible to compare the effects of all these things, the only thing is you must understand what you must change to achieve what you want since there aren't really bushings in the suspension, but only beams that make a simplefied geometry with a singular stiffness value

 

Ah thanks, that confirms pretty much what I guessed from browsing through a few vehicle files and not seeing anything in that regard. Perhaps I'll try introducing some quite small beams at the chassis-end of the existing ones with very high strength and very low stiffness at the end of the existing beams? I suppose that should be possible, even though it might not be terribly accurate - I haven't got the experience yet with the jbeams system, perhaps that sort of thing in general is asking too much of it. But it would be fun for me to try a) a racing car with much increased chassis stiffness all round, b) such a chassis with "rubber" elements added; c) changing the suspension geometry for better and worse, using the very stiff chassis at first, then adding the "rubbers".

 

you can't do that since 2 beams require a node in the middle and that node acts as a pivot point, but you could use bounded beams to have them be very soft until the compress or extend a certain length and then return to the "steel" stiffness

 

Thank you both, I'll play around with that concept then; sounds quite close to what I have in mind.

That is, umm, of course rather interesting to hear. So if I just kept adding strength or additional beams to a model, what will happen in the game? Will it still be usable or break at unexpected points, or become uncontrollable in like rough terrain? I mean, keep in mind I'm not so much interested in dropping these cars off 200m cliffs as I am in finding out what happens to a single car in a fast corner, with dynamic changes and slightly uneven road surface etc.

Guess whatever the physics instability is, I'll run into that - what I'm trying to do is obviously going in the direction of creating a ultra-stiff chassis. I'm under the impression that most "normal" racing sims either don't take that into accout at all, or at most use a simplified model such as a torsional spring between the f/r axle fix points - which I suppose has a lot to do with the perceived differences between these conventional ultra-stiff models, and the driving feel in BeamNG.

 

Let's just say that when you run into an instability, you'll know it well before you can even attempt to drive the car lol

imagine this https://documentation.beamng.com/modding/vehicle/intro_jbeam/common_issues/Instability.mp4
but way more extreme

 

Ah right, I did notice actual breakage when naively decreasing node weights already - the car's suspension simply broke when taking a kilogram of weight out of the wishbones. Lots of ways to figure out for me, I suppose on the "rigidification" front right now with cars like the Bolide it might make most sense to use the virtual strengthening bits, which have more reasonable spring rates and springs (somewhere around 100-600kN/m, probably not in danger to start acting weird) to rigidify the entirety of the car, right?

Edit: It worked for a bit and I think it made the entire car feel more responsive and controllable, but as soon as I took the front rigidification over a certain limit, it starts snapping tierod ends at spawn - probaby the jolt at spawn overloads the now rigid system. I'm starting to believe, my experiment ideas might require a whole lot of RTFM...