I'm currently trying to make a race car and have given it 100% anti-dive and anti-squat suspension geometry, with the SVSA ICs (side view swing arm instant centres) for both axles being at the centre of gravity. I'll deal with the wheel hop ramifications later (as per this thread), so I'll likely dial it back, but I do want there to be some anti-geometry remaining due to future high downforce requirements. (And it's not like the pro drift BX is any worse off with the maximum anti-squat available...) For now, though, the rear suspension still noticeably squats for a bit when torque is sent through the rear axle, compared to the front suspension, which doesn't dive as much under heavy braking. I feel like I'm somehow doing this wrong, since the coilovers on both axles are already running Scintilla spring and damping rates on a car half as heavy. The only theory I have is that the suspension mounts being set well below the rear wheel hubs are to blame (compared to the relatively high-set front suspension mounts), but I have no idea how to fix that if it is indeed the source of the issue, since moving them up isn't really an option. (P.S. The attached test car is named "[Classified] [Classified]" and should be at the beginning of the car list.)
I guess I'll try moving the rear IC forward or backward and seeing if that does anything... (Update: moving it to roughly where the front axle was didn't do anything in the slightest.)
Update: no change in initial squat after moving the rear IC backward. After some more testing with less grippy tires, it seems like the front does dive to the same extent. What's even more frustrating, though, is that stiffening the springs tenfold didn't have any noticeable impact on the squatting and diving behaviour. I think I might need to add more stiffness to the bellcranks and pushrods, but looking at the Formula Ibishu and Carbonworks F4 mods, they run much lower bellcrank and pushrod spring values (e.g. 4,001,000 vs. 100,100). Any help from those with more JBeam experience would be hugely appreciated, as my motivation is starting to peter out.
Well, I'd say that 100% anti pitch geometry won't simply "delete" pitch movements altogether. Usually, anti pitch geometries increase the stress on the control arms, which can lead to compliance. Maybe try to add a coilover connecting both bellcranks so you can make it harder for both wheels to compress, while keeping the rigidity at each wheel not too high, like the Triplex system on Koenigseggs. Although, that system would require stiffer ARBs, since it is a pro roll system by itself
Quick check shows me your IC still is below the COG line, making the anti squat not 100%. Attached is something i eyeballed to be pretty close to 100% antisquat, and ideally you'd like the front IC to be at the COG line too for the chassis to stay straight. But personally i'd dial it down a few notches an use the setup Xupaun posted. Front does lift slightly on acceleration, also the way other things are set up will cause it to rotate the body backwards slightly, for instance wheel inertia change and the torque arm(how the torque driving the wheels is applied to the chassis).
I totally agree. I spent the last 6 years of my life designing and building Formula Student cars, and we never used Anti pitch geometry. It brings unnecessary stress to the Control Arms with little to no advantage. I'd personally stick with up to 50% anti dive and squat and using other methods to increase the "pitch rigidity". A weird option would be adding anti pitch bars, that behave like ARBs but connect the wheels from the same side, instead of the same axle. Anyway the extra coilover/spring can add the tunability factor to that suspension
Although the altered geometry you provided for the transaxle yielded no changes on its own, redefining the front and rear torque arms (in particular removing the "torqueArm2" arguments and setting their nodes to the standard "torqueArm" argument) in the wheeldata parts definitely did. It seems like the "torqueArm2" beams were being defined along the halfshafts, and thus actively caused rise/dive/squat etc. in a way that made any anti-geometry all for nothing. I was worried I might have to create lighter wheels specifically for this car as the touring Pessima's centre-lugs helped a little, but thankfully that doesn't seem necessary anymore. In fact, it actually seems like your setup results in negative anti-squat now Thanks for your input regardless! I think I'm ready to dial it back down to fix the wheel hop and whatnot.
you can see in the graph i provided that the rear IC slightly exceeds the COG line, making the anti squat exceed 100% and thus indeed causing a lifting force in the suspension. It reduces the squatting from your setup drastically although not fully eliminating it due to the way the torque arms were setup. It was merely meant to show you that there was something wrong in your anti squat calculation approach so I didn’t put much thought and effort into perfectly getting it to 100%. I highly suggest sticking to the advice Xupaun has given, 100% antisquat is a bad idea. What type of racecar are you trying to make? And what era?
I'm well aware. Although the math in this article still goes over my head, it did help explain the side effects of aggressive anti-geometry: https://www.waveydynamics.com/post/anti-geometry Without saying too much, it's a replica of a 21st-century concept car, which has an established rear pushrod suspension setup with the two coilovers located roughly parallel to each other and above the transaxle. The intended performance target is similar to Group C cars.
