Springs, ARBs, and Rebound

[BrodeurSux]

This is my fisrt day on Muscle Inc. hello all, for starters. i apologize in advance if this post is in the wrong forum....

i've got an INGS RX7 (S998) that i've been running at Suzuka. normally, i'm running 1:56s on a good day. got the build i like, i think most people are running this one. as usual, playing with the tune, trying to figure stuff out.

meanwhile, i'm reading Tune To Win and i'm in the section about lateral weight transfer and body roll.

Carroll Smith says body roll is bad, and the best way to reduce roll is stiffen springs, and stiffen ARB's. so i'm like, 'okay, screw it'.....

i take my RX7, and using the 'just play around with it' philosophy, i stiffen both front and rear springs to the max, stiffen both ARB's to the max, and while i'm at it, i turn both rebounds all the way up to 20 (because so many LB tunes have very stiff rebound, and very low bump, although i'm still not sure why....but what the hell, i give it a shot)

you know how you know when you stumbled across something you did right (i'm sure you all can relate), because you go out and run one lap in tuning mode, and the car is faster? i immediately start ripping off 1:55's. make some minor tweaks to the camber and front toe for better turn entry, then i toe the rear wheels in for a little better speed, and when i'm done i'm left with a PB of 1:54.8xx, and i feel (as usual) like i left some time out there because there are still two turns on that track that kill my lap times.

The thing that's frustrating about this game is there seems to be no rhyme or reason to what works from car to car. i'm a math guy, and things need to make sense to me. there is no sense as far as i can see when it comes to tuning. there's no pattern, or formula. frankly, it frustrates me, because physics are laws, and shouldn't vary from car to car. but i know this is a video game, so....

what do you guys do with springs, for starters? establishing a starting point for spring stiffness is always an aribtrary endeavor for me. logically, springs stiffness (front to back) should be set according to weight displacement - a front engine car should have stiffer front springs where a mid engine car should have stiifer rear springs. and static weight or mass should also be considered, a 2,000 lb car would have softer springs than a 3,000 lb car.

but then there's my RX7, super stiff everything (springs are like 637 lb/ft) and the car only weighs 2,000 lbs. yet it's faster than when i had my springs set around 300 lb/ft.

anyway, any feedback would be appreciated.

[Fit4aking]

For starters I run all cars stock to determine its tendancies.  Then I figure the original SWR and Bias and adjust as necessary.  Lots of good eye-crossing reading here:

http://www.muscle-inc.net/index.php/topic,12.0.html

Depending on what the car tells me I adjust for Bias first then SWR second, too bad each car is different.  If it feels soft I'll SWR first but with most cars I try to tune it out with bias before larger changes.

[fndrbndr]

I'm just starting to get away from the spreadsheet approach, myself.  That said, blooze's thread is as good a starting point as any. 

[Hoplee]

I too wonder why the cars all appear to be so different. I don't think FM2 has actually modeled a hundred different suspension assemblies that it then has to keep up with during races, I suspect that there's something very simple behind the scenes that's generating a ton of variables.

I've made several spreadsheets based on several theories, but have reached a point where I tend to go back to defaults + very minor tweaks. I think that in the beginning, I was trying to tune my way around my bad driving habits!

[Snyder005]

Have you looked at some of the Exige set-ups?  Max ARB's, Max or Min springs, Max Rebound, Min Bump.  Crazy stuff like that seems to work somehow.  Like the 1/40 ARB settings.

[Hoplee]

One loose string I always wanted to pull on is "ride height"

Tuner versions of cars not only have different defaults for ride height, the have different ranges, with some versions offering a wider ultimate range of adjustment than others. Maybe, just maybe, there's a common suspension model that they f with the orientation of to generate different types of response...

PS: The "B Class Missile" ferrari 250GTO has really weird settings that happen to work, for reference. I like using it as an example because it's such an extreme.

[BrodeurSux]

Have you looked at some of the Exige set-ups?  Max ARB's, Max or Min springs, Max Rebound, Min Bump.  Crazy stuff like that seems to work somehow.  Like the 1/40 ARB settings.

tell me about it, my brother runs that freaking Exige Cup about 85% of the time. he runs great times at Laguna, Mugello, Suzuka, but i wanted to kick him in his butt when he ripped off a 1:43 at Siverstone Grand Prix! he got that exact whacked out set up off the LB and sadly, it produces fast laps. although i will give him credit, he did tweak that tune quite nicely before posting his Silverstone time.

as for the 1/40 ARB setting, it's not that crazy in an AWD car. i've heard of guys who tune Imprezzas and Lancers in the real world that actually remove the front sway bar to eliminate the understeer. I've got an AWD Lambo LP640 with a 1/40 ARB set-up that's nicely tuned and i've put up some good times in it.

That said, blooze's thread is as good a starting point as any. 

i got wind of the 'bias' approach to tuning from someone from here that posted a nice thread in the FM2 'racing techniques' forum. can someone point me in a direction (a link, perhaps) on where i can learn how to tune using this technique. i'm a tad bit thin on the terminology.

[BarbecuePete]

Hey Brodeur, welcome to M-Inc, Im fairly new around this place myself too, but all the old hands have been really helpfull with advice and tips...

Here's a link to the thread I started asking for advice when I was starting out here, there's a link to a couple of spreadsheets in there, I use the Tonka spreadsheet for my tuning which i think uses older calculations but it works really well for me, I think the 'Kitchen Sink' uses the latest calcs...

http://www.muscle-inc.net/index.php/topic,772.0.html

[fndrbndr]

Honestly, I've been using the Tonka Toy as a starting point, though I'm finding that the default stiffness and 54.5% weight bias really aren't optimal for me, even on FR cars.  I usually like stiffer springs, at the very least, to the point that stiffening to about 78% is worth a half second at medium to long tracks.  The Tonka Toy is also a bit more accessible.

[Ske]

what do you guys do with springs, for starters? establishing a starting point for spring stiffness is always an aribtrary endeavor for me. logically, springs stiffness (front to back) should be set according to weight displacement - a front engine car should have stiffer front springs where a mid engine car should have stiifer rear springs. and static weight or mass should also be considered, a 2,000 lb car would have softer springs than a 3,000 lb car.

I've tried many approaches during my Forza career but in the end I've settled on a set of numbers that I use across all my cars, regardless of weight. For a quick hotlap car (non RWD) these days I'll most often go really really soft (200-250 front springs), but I might increase that by 50 or 100% for a more stable online racing car, depending on track and class most of all. My FR starting point is 350/200. I'll then have to analyze per car if the spring rates work.

To make a long story short, in order to reach the highest lap times I've found no relation between vehicle mass and optimal spring rates. There is, however, a relation between spring settings and suspension configuration - but since we're not told what kind of suspension we're dealing with (strut, arm, wishbone, "log") that's much down to trial and error.

A lot of people are sticking to the idea that front biased cars should have front biased springs, which is fine for road car safety and wear and tear - but when it comes to tuning a race car, those conventions are in my opinion not that valid. For example: most race FWDs and AWDs in the real world are set up for massive oversteer with beefy rear springs and ARBs - and that also works in Forza. Like you said, 1/40 is not an unrealistic ARB setup until we know exactly what "40" means. Many RL race cars run with just 1 ARB to reach the desired amount of over- or understeer.

The thing that I think makes the crazy Elise setups work is that stuff like feedback, bottoming out, bump steer and wheel hop (among other things) don't seem to be modelled well (at all?) in Forza. So, the most important thing to consider when setting up your suspension is the rate and severity of load transfer.

When it comes to dampers I'm still fumbling a bit, but I've come to the conclusion that a 4:1 ratio of rebound:bump seems to work well. The "standard" 3:2-2:1 setup still works, but in most cases I've found it not as effective - especially in AWDs and MRs. Again, this is not as insane as you'd think from looking at road cars. Race cars seem to run "unconventional" damping rates quite often.

To conclude, the major point I'm trying to get across is that setting up a race car can be completely different to setting up a road car. Since comfort and wear is not an issue we can go to extremes to get the cars to be as quick as possible. There are shortcomings in the Forza physics model that cause certain outrageous setups to work (re: Elise) but in general it's not as far out as you might think. The exception might of course be that the lack of consequences from bottoming out and the absence of wheel lift means that soft springs are favorable.

I think there are certain cars that are "broken" like the Elise, but I tend to shy away from those 

[BrodeurSux]

When it comes to dampers I'm still fumbling a bit, but I've come to the conclusion that a 4:1 ratio of rebound:bump seems to work well.

it's ironic you said that, because i've come to the same exact conclusion. i've noticed that when i keep my bump at 25% of rebound, the car feels good to me, although i think it may just be a personal preference for you and I.

damping is probably the most difficult aspect of FM2 tuning for me. i'm still experimenting with different settings. i think the thing with soft bump is the car seems to hold traction better in long, fast curves. couple that with soft springs and stiff ARBs, and you kind of have the best of both worlds because, as you (and others) have said, the game doesn't puninsh you for bottoming out, which in IMO the biggest flaw with the physics. if cars bottomed out, and lost control the way they should (i even suggested they should build partfailure into the model for FM3) set-ups would change drastically.

but as such, soft springs (in FM2 as well as the real world) give you added grip, while stiff springs give you responsiveness at the expense of traction. but add stiff ARB's to give you the roll resistance you need, and i can see why a lot of cars utilize this type of set-up.

[Williewanka93]

Just wondering, can someone put a link for tune to win

[Hoplee]

I'm certain that the game DOES punish you for bottoming out. It might be that the effect is counteracted somehow by the ARB, damper and/or spring settings, but there is definitely an effect. 2cents

[Snyder005]

how can you tell if you bottom out?  that's one thing i never understood, or recognized while racing.  I know my Renault that i posted here is wicked fast, but it breaks traction like mad if your not careful and i don't know how to fix that.

For me, usually i run soft springs, with very high rear rebound settings and low bump settings.  I'll post up my 1969 Camaro SS set-up in a bit.  I also tuned a missile Ferrari GTO for the HLC, and sent it to Jczero, who ran a top 50 time for that week with it.  Both cars follow this principle for the most part.

[Ske]

If the bars in telemetry>suspension turn red it means that you've bottomed out. With the typical "leaderboard" setup, your fronts and outsides are probably in the red quite a bit during braking and in the first half of corners. It doesn't really matter though, Turn 10 seems to have forgotten to implement the consequences.

Hop, what you might be noticing are the effects of camber change - camber has a HUGE impact on handling in Forza, and for some suspension configurations (struts etc) the camber change during suspension load can be very significant. Certain cars like the E-type, Lexus IS, Astra, Neon and Golf mk5 benefit from harder springs, dampers and/or greater ride height to minimize this effect.

I don't know if you remember my FCT E-type running -2 and -1.5 degrees of camber? This was to counteract that, but when revisiting that car I found that by toughening up the springs and ARBs and increasing the tire pressure a bit I could lower that to more sensible amounts.

Now, -2 degrees is not unheard of in the real world but like I mentioned Forza seems to exaggerate alignment effects, maybe as much as 4-5x when it comes to toe. In real life a race car could easily run 2-3 degrees negative camber and 1+ degrees of toe-in but in Forza, more than .2 degrees of toe or 1 degree camber is usually enough to seriously compromise a car.