subbed (and bumping an old thread) - interesting discussion as I research moving spring rates

 

I'll add my bit here. Too lazy to read 6 pages.

I was also from the camp that prefer larger spring rates in the rear, which is contrary to all motorsports level suspension enginering.

Then i started playing with my rake and jesus christ. 6mm more rake made the difference between a balanced setup at 8/9 spring rates and rear sway at medium, to a setup that is wildly uncontrollable in oversteer. I moved my rear sway to full soft and reduced rake to control it a bit better

This gives me a hunch that perhaps a spring choice based on corner weights would be best, using rake to manipulate your roll couple and control load transfer front/rear. This deserves serious experimentation.

As far as other bits on roll centers and such, increasing your front grip as much as possible is the most important thing you could possibly do. Rear grip is easy to find, front grip is not. I've lowered my evo more than people reccomend and measure 6.3 in of ride height at the pinch seams. I can't lower anymore without my 295's contacting the front fenders, and i currently run -3.5 degrees of camber up front on 8/9 kg KW's. But before i start messing with front roll centers, i think i need to investigate running extra rake with lower rear spring rates.

 

^ how are higher rear rates contrary? The evo motion ratio has been reported anywhere from 0.6 to 0.8 so it's not an effective higher rate - more info in the how to pick your coilover springs thread

 

I am pretty bad with suspension but...

The rear is much lighter than the front and as such does not need as much spring rate to control roll. If you have comparable compression damping front and rear, the rear effective spring rates can be quite a bit less.

The rears are in compression during corner-entry, mid-corner, corner-exit and on bumps. Bumps are handled better by the dampers with lower spring rates. You will get more front grip with softer rear damping and spring rates during corner-exit (squatting). During mid-corner you are just controlling the roll and as stated above, less spring is needed for the rear. For corner entry, you are trying to keep your *** on the ground and control the roll.

You need to be able to control the transition from one state of the corner to the next. Having a very "tight" car will have minimal transition from one corner state to the next. Having a very "loose" car will have additional transition movements associated with it. Personally, I try to drive as "loose" as I feel comfortable without overshooting the corner transitions for the particular track. IE: it would be bad to have a "loose" car for S curves where the quick transitions would upset a "boaty" car.

I am having a hard time conceptualizing how rake impacts rear grip. Albeit rake certainly helps the downforce of the car.

I would really like to get a set of dampers that I could re-valve/re-piston in my garage to test some of these ideas for different tracks beyond just increasing/decreasing the slopes.

 

The rear suspension has a slower motion ratio than the front. So the effective spring rate at the wheel is lower than the spring rate of the spring.

 

I am aware of that. I was addressing why having a lower effective spring rate in the rear may not be as outlandish as it is portrayed here.

 

On an X, the front MR is about .96 and the rear is .84. A bit higher than the 8/9.

The common school of thought is to choose spring rates based on a natural frequency which requires you to consider weight distribution. Since our cars are so light in the rear, this would mean all of those first coilovers that hit the EVo X market were near the money with their advertised 8/6 spring rates and such with this approach.

The difference here is that our multi link rear suspension is much better at controlling the camber curve and roll couple than a mcpherson type setup. This forces us to pick way stiffer spring rates and stiffer sway bars to REDUCE rear end grip to achieve a balance we want.

What i think i'm finding out is that you don't need a stiffer rear end to achieve the balance you want, you just need to manipulate your roll couple. Raising the rear will keep the RC and CG closer together, forcing that side of the car to load quicker. If you exceed the load of the rear tires sooner than the front, you get oversteer. This approach is done through geometry. The previous approach is one of brute force via springs/bars to achieve the same roll rate per LB of force added to the outside tire. One of these approaches will be much more comfortable and compliant than the other, and theoretically, both will have the same ultimate lateral grip.

So, if you don't like your vehicles balance, perhaps first you should play with ride heights and ball joint heights rather than springs.

 

Not to throw a wrench in this discussion, but i'm finding more and more motorsports teams these days completely disregard ride frequencies of their vehicles. Their choosing stuff that simply shouldn't work given the old school of thought, but it is obviously working well. How does that apply to us? Perhaps we shouldn't place as much emphasis on ride frequency as we do. It's an arbitrary number to shoot for, and motorsports teams are doing well dismissing it in their design criteria.

I suspect the optimal approach might be a mix of the two. I'm purchasing new coils in the next month and i'll be starting with a Square 14kg/mm setup. That should give me enough spring for 335 hoosiers

 

Motorsports tems may be working with a car that has significant downforce, and/or a car that has proper front suspension (an SLA dual a-arm setup), and they have even done other things to optimize geometry. You can't just look at spring rates on some one elses car and "I should replicate that" without knowing what else is done on the car.


And yes, the Evo X has a different spring requirement from the 8/9.

 

I'm not saying we should replicate their setup, i'm saying if they don't consider ride frequency, why should we?

I've seen open wheel non downforce cars, open wheel moderate DF cars, open wheel light DF cars, closed wheel non df cars, closed wheel moderate DF cars, blah blah blah blah blah. Basically, i've got enough data points to consider my observations as enough to warrant totally rethinking how we approach things with our 4 door grocery getters. None of the properly engineered examples i've seen in the last 8 years have considered ride frequency in their spring rate choices.

 

Ride frequency is typically addressed for ride quality. If you're aiming for a certain balance in the car, ride frequency goes out the window, unless you just want to compare the front vs the rear.

 

Natural Frequencies are just a way to explain how stiff a spring mass system is. Super basic, but it certainly can tell you if you have one end catastrophically stiffer than the other, are far below or above a reasonable range, or to compare car to car stiffness. What it doesn't do is give you any idea how the effects of lb/tire section, transional response (due to roll couple or shocks), bar rates, chassis stiffness, diff crosstalk (left/right, fore/aft), etc.


I can see teams not concerning themselves with NF on proper race cars and smooth tracks but for production cars with usually different front/rear suspension types and definitely geometries and weights, its a good start to defining the system and a good starting point.


After modeling up what happens when you lower the front of an evo, I can definitely see your rake change having the effects you're describing. From stock to -2.5" right height up front lowers the front roll center 6-7" if my numbers are right. I haven't modeled the rear, but being multi-link it shouldn't have nearly as severe of a move (Upper arm brings instant center inward, Strut doesn't so once past vertical the roll center drops very rapidly).

 

my current setup is quite a bit "off" from what everyone is recommending, but the car drives quite neutral, though i admit the tires may just be gripping like crazy.
i plan to drop the rate on the front to get a bit more rotation. it seems to work for me.

i wonder what all the fancy math would tell me, if i knew how to use it LOL.

i think i found some info on why my car does decently even with the amount of roll i have and the softer springs. also, my car isn't really all that low:

 

All good stuff Kaj. I guess my point was spring rate choices are only half of the equation. Ride height, or more specifically, roll couple must be considered if you want to achieve a certain balance.

 

14kg/mm square will not be enough for 335's.

If you want to know why rake changes have such a noticeable impact on the balance of the car, think about what happens to the front or rear camber when you change the ride height. Turns out if you make 'camber neutral' changes in ride height/rake the change in balance is no where near as dramatic.