B pillar deformation and side impacts

The issue specifically with the B pillar is something present in all vehicles, however best demonstrated with the sunburst. The B pillars of vehicles in this game have a little bit of inaccuracy worth looking at. They don't quite deform right, and it appears that the top of the B pillar is simply held in place too rigidly relative to the vehicle's roof. All vehicles where tested with a side impact 'test' to demonstrate this. Doors removed, however it doesn't matter as the coltris seem to spread the load of the pickup truck approximately the same way doors will. I drove the D15 into the side of the vehicle at 40mph, which is something akin to what the IIHS side impact test is supposed to replicate.



The other thing I must point out, 2 door vehicles have some mesh scrambling issues in side impact tests. The grand Marshal's roof collapses in an unrealistic way during the same test.

 

I noticed the roof collapse issue after 3.7.1 or 2. It was pointed out tweaks were made for better A-pillar deform and I really think gabe instituted
some of the tweaks we (you and I specifically), had made?. There is a compromise somewhere with A-B pillars. Too much adjusting on the B pillars
stops the roof from "tenting" in a broadside collision which I personally can't get enough of.

 

That migth very well be the case. I'm thinking that if the 'corner' beams between the B pillar and the roof are made much much weaker it would allow the roof to tent during a broadside, simply because it would allow the angle between the roof and the pillar to change more easily during a crash. I'm also thinking that putting a single node behind the B pillar and creating an actual solid pillar, rather than a column of nodes supported off the floor, would function far more realistically. Such as this image, and the blue colour doesn't mean much. I just wanted to differentiate between 'supporting' beams and the main shape of the car's cross section.



- - - Updated - - -

Sussed! I got it to deform better by carrying out the changes I illustrated... plus some more tuning. It's now acting like it's own solid structural member... which it should. It's not quite there, I suspect it needs to be further rigidified. The top of the pillar no longer holds itself perfectly rigid relative to the roof which means that in side impacts there's roof deformation concentrated in the middle where the pillar has forced the roof to buckle.


Also, it REALLY tents up now




Here, I've tagged all my changes with //edit and some explanation. Is that much better? Try it out, see if you can use it to get the GM to play nice too!
View attachment sunburst.jbeam

 

Will check this out. To be honest I haven't done a single beam tweak on the Sunburst yet but that's about to change. Are you using Masa's B/N editor?

 

Not yet, I just use notepad++. I also redid the Hatch, modifying the B pillar and leaving the rest as Gabester left it. Seems to work better than my previous one.

View attachment hatch.jbeam

again, all my changes are marked with //edit.

 

Did Gabe institute some of your refinements? I looked at the hatch file briefly awhile back and noticed the official file had what appeared to be
the same beam additions you had done in your file. Appreciate the highlighting, makes it much easier to see changes.

 

It's possible, though to be fair there aren't a massive amount of approaches to this that I know of. however he knows enough to tune the beams perfectly for this, so he's done a better job.