mazdaexe

Hi all, I have a set of teins flex w/ EDFC. I was at a autox practice last weekend and was trying out my settings. I decided to use a 4F 4R setting on the EDFC (0 hardest, 16 softest) after several runs. My car tends to drift out a little bit in the middle of a sweeper, then a friend told me to set my rears a little bit softer. So I set to 4F 6R setting and sure that helped a bit. but im trying to understand the physics behind it....
Anyone?

Ophitoxaemia

the physics isnt as complicated as the effect on handling, which i find pretty complicated!

stiffer shock settings cause weight to transfer faster at those shocks.

ive also heard it said that compression (bump) setting controls the unsprung weight and the rebound controls the sprung weight. so if the car feels "floaty" it might need more rebound. if the tires spin the instant you touch the gas, there might be too much bump.

there are some web sites with more description. typically once you set the bump correctly, then you can tune the car with the rebound. on my autocross car i use a very soft bump and rebound setting at a very slippery site to give me superior traction. on concrete i will use stiff settings to make the car easier to drive.

james

shelleys_man_06

Physics eh?

Okay, here goes...

Your spring/shock combination is going to follow a 2nd-order ODE, which goes by

mx[2-dot]+bx[dot]+kx=0

m = mass
x[2-dot] = acceleration
b = damping coefficient
x[dot] = velocity
k = spring constant
x = position at time t

Dividing by m, and doing some crazy math, you'll wind up with

x[2-dot]+2(zeta)(omega[n])x[dot]+(omega[n])^2x=0

where

zeta = damping ratio = b/[2(sqrt(km))]
omega[n] = natural frequency = sqrt(k/m)

This equation describes free vibration with viscous damping.

In most cases, the system (spring/shock) is usually underdamped. I believe the damping ratio rarely goes to unity (1).

For the underdamped case, in which 0 < zeta < 1, the response function is

x(t)=exp(-zeta(omega[n])t)*{[(zeta/(sqrt(1-zeta^2)))*x(0)+(1/(omega[d]))x[dot](0)]sin(omega[d]t)+x(0)cos(omega[d]*t)}

If the initial velocity of the system is zero,

x(t)={x(0)/(sqrt(1-zeta^2))}exp(-zeta(omega[n]*t)*sin{(omega[d]t)+arctan[sqrt(1-zeta^2)/zeta]}

This probably means nothing to you, but this is the function for what is happening when your suspension goes under some dynamic load. Pretty scary isn't it? There are many things you can actually find here, given a significant amount of useful variables. The damping coefficient is very important in understanding the time at which it takes for harmonic motion to reach steady-state equilibrium with its surroundings. Also not that omega[d] = omega[n]{sqrt(1-zeta^2)}. This is the damped natural frequency. Basically, for racing purposes perhaps, it's important to keep your damping ratio as close to unity as possible. You can play around with the equation, using values for b and k, to get the damping ratio experimentally. My hands hurt too much to write about it.

Zaku-8

hey shelley's man, do you have mathtype or something like that?

I'm not engin or physics... so its hard to make sense of those equations, even though when in their symbols i think i could probably understand it

shelleys_man_06

Does Mathtype work in forums?

If you're ever confused about the mathematics, just pm me. I would be more than happy to explain it more in detail.

mazdaexe

er.... thanx shelleys_man_06?!

how about simplify things and put it in plain english? thx

actually. after looking at ur formulas more carefully, how do I factor in front and rear damping force seperately?

RX8_Buckeye

shelleys_man_06: nice job of describing the suspension system in terms of a simple harmonic oscillator. However, I want to point out that the 2nd-order ODE applies purely to rigid-body motion of the vehicle body in the vertical direction, and does not describe relative motions of the front/rear or left/right sides of the vehicle. The handling of the vehicle is also affected by the properties of the powertrain mounting system (which can be thought of as another mass-spring-damper system), and the body flexural properties.

Basically, the simple 2nd-order ODE you presented is great at describing the basic functions of the spring rates and shock absorber damping, however it is not at all adequate for describing the handling characteristics of a vehicle as fine adjustments are made to the front/rear suspension components. A more involved model is required for this purpose.

shelleys_man_06

Well, I had only meant to describe the vibration motion of the system, not the overall vehicle dynamics. To do so is way more complicated, and out of my realm of knowledge.

Vehicle dynamics involves much more than describing a free vibe with viscous damping.

Anyway, don't hurt your head in this area. The ones who design these products probably already did the sim modeling.

Oh yeah, bong, if you're ever confused, I can write up a Word document that will explain things easier.

mazdaexe

Thanx shelleys_man_06 for your time and effort. Even you sent me a detailed word document, I would still be confused.

It would be easier if anybody could just tell me setting the front firmer/softer = ...
and the rear firmer/softer = ..... and why..
thx!!

Zaku-8

i was under the impression that mathtype could output jpegs. maybe not though, its been a while

r0tor

iTrader: (1)

the problem with the 2nd order ODE is it doesn't count sway bar and tire sidewall effects and assumes that there is an ideal value for dampening... which there isn't and why all suspension tuning is still being done predominately by feel.

shelleys_man_06

My sentiments exactly.

Your best bet is to take your car to the track, and make adjustments accordingly. Doing the math is time-consuming and pretty much worthless.

Suspension tuning is what I consider a black art. It's difficult to understand fully without experimentation. You can only simulate crap for so long.

Dark8

I was never into math. Are you on stock tires? If so start at the softest setting and see how far you can go before you start loosing grip. If you are on R rubber then start with a little firmer damping. I did a quad auto-x (4 in 2 days) last weekend and with my JIC's I reached a point like you where I was too stiff for the OEM gumbals Mazda calls tires and was drifting around corners. I backed both off a few clicks and it went away. I know we are talking apples (Tien) and oranges (JIC) but my fronts were six clicks stiffer than the back. I know this wasn't based on a mathematical equation, but I hope it helps.

GeorgeH

iTrader: (2)

I'll try to explain the physics in layman's terms, since I’ve been attempting to evolve my understanding of this subject over the past few years. You guys tell me how I am doing.

Since you were experiencing loss of grip in the middle of a sweeper, we are talking about steady-state, as opposed to transition. But dampers don't affect stead state grip, right? Wrong.

The reality is - your suspension is never at steady state - it's always working, even when you are driving in a straight line down the road. The wheel is always hunting, moving up and down, and the suspension is always trying to keep the tire from losing contact with the ground. Same thing while you are in the middle of the sweeper. But the tire isn't just moving up and down relative to the car, it's also moving up and down relative to the road.

How can the tire move up and down relative to the road and not loose contact with the road? Easy - the sidewall of the tire deflects (compresses & extends). So, the center of the wheel move up and down relative to the road and relative to the car at the same time. So, relative to Shellys_man's post, it's actually a 2 DOF 2nd order ODE - there are two degrees of freedom (the tire sidewall flex and the suspension movement).

It's this fluctuation in the sidewall deflection that causes your car to loose grip over rough roads. The more the tire deflects, the greater fluctuation you have in the tire's reaction force with the ground. And the reaction force with the ground is what defines how much grip you have.

Chassis engineers call this "tire force variation." Their objective is to minimize it. How do you minimize it? Well, you keep the unsprung weight low, carefully pick spring and sway bar rates, and, carefully tune the dampers. Dampers can play a big role in tire force variation. Too much compression damping and the tire skitters over bumps - you have too much compression stiffness. To little and you can also give up grip, or transitional response suffers. Top-tier race teams spend big bucks determining the optimum damper settings on a track-by-track basis.

So, when you softened your rear dampers, it changed how the suspension responded to the fast fluctuations of the wheel as it rolled over surface imperfections. For any given corner, there is a "sweet spot" for the dampers that maximize grip. Sounds like for the conditions you experienced that day, 4 was too stiff, so 6 allowed you to pick up grip.

As others have stated, in the end you just have to experiment to find out what works, unless you have the budget of a F1 team.

Ophitoxaemia

my understanding is that dampers provide their greatest contribution to tuning on corner entry and exit where the car is making large changes in weight transfer. this assumes the dampers are good quality, are not overheating and are in roughly correct adjustment for the spring rates and weight of the car. they control how fast the weight transfers given the braking and steering input of the driver, and this controls under/oversteer in transitions.

i have some results from testing my race shocks on a shock dyno. it is pretty interesting- there are lots of effects to look at, which as how well the shock controls slow versus fast shaft speeds, or how well it handles a combination of slow shaft speed (weight transfer changes) with high speed (bumps). so its not just a single number or three to describe a damper.

what ive learned from racing is what you drive on and with the weight transfer. so dampers are very important! fortunately the rx8 features very good dampers stock.


james

shelleys_man_06

Yeah. The damping coefficient is attatched to a time variant derivative, so it would be easy to tell that the system is transient, hence the 2nd order ODE.

Like I said, there's too many variables to keep in mind. Don't think, just drive.

Dark8

Remember that tires are also a limiting factor. With stiffer suspension, tire load during weight transfer happens much faster than with a soft suspension. Tires with more grip will handle that transition better. If you are on stock tires, a set of wide BFG KDs, Kumho MXs or race rubber will let you run more damping without loosing turn-in or mid-corner grip.