Suspension setup guide.

what kind of formula i would need to determine travel based on spring rate ?

 

You would base it on graphs like those you already posted.

 

so basically going through all 158 of those spring rates.

 

No, not at all.

You said it's not linear, but I bet it follows a curve/pattern. You can pretty quickly get an idea of that curve based on springs, ride height, and packers (can't quite remember the interaction there; some tests will indicate) and create a formula to approximate it. Only a handful of data points will be needed.

This in comparison to doing 2 tests (stiffest and softest) and taking the average, which you said you are doing.


It's not laid out on a plate for you. But you can bet people driving these cars competitively (even at league level) put in some time to do this. It's part of gaining an advantage over your rivals.


@Simulation_Player here's a hypothetical for you: Let's say tomorrow a new channel got added (by the game, and a telemetry plugin obviously) that shows when you're on the bump stops. It says 0 when you're not and 1 when you are. How are you going to use that data in tuning your setups? Will you aim to have it always 0? Or always 0 when cornering (define 'cornering'... wheel angle? Lateral G force threshold?)? Will you completely avoid 1 when cornering at all, or will you accept it when hitting certain speeds where aero takes over?

What you are likely to do is not just rely on "am I on the stops or not" but "how much am I compressing the stops, and how much am I willing to at this moment?" To do this sort of analysis you'll need to know where the contact point is, and you don't just do that live - you'll run some tests, find where 0 changes to 1, set that as your contact point in Motec and then define a function that tells you how much the rubber is compressed. Then you'll aim to keep within certain compression limits based on other factors (speed, cornering, ...).

So the step you skip in that point is having to run those scatter plot tests to find the contact point, but from there it's the same: define/estimate the contact point based on current setup, and then run your analysis from there. Even then there's no 100% perfect way to set up your car - you theorise, apply, test (drive), and occasionally recheck and experiment and alter your theory based on what you find.

 

i tried a way to estimate this curve, not sure on my method tho.
basically i ran test on lowest spring(base) then from base +2 clicks ,+4 clicks and MAX value.
i looked how much travel i lost at maximum spring rate (iirc it was around 12mm).

some extremely simple math like >

if 32 click = 12 mm lost
then 1 clicks = ? mm lost
(12 x1) /32 = 0.375 mm (~0.4 mm for every click)

but then at +2 click and +4 click test it shows i lose 2mm every 2 click (so 1mm every one click).

I don't know how i can form a curve out from such tests, any ideas ?

 

since bumpstop are nonlinear is it no problem to ride them a bit...hence initial bumpstop hit is much harder to detect while driving. So giving a channel that shows when bump is hit IMO solves bigger problem.
After than i can just look at how longer the bumpstop is activated, how much suspension is compressed after it has been activated and making adjustment then will be much easier. IRL teams don't have to go through all these unnecessary troubles, neither do people in other sims.....it is not big ask to atleast add a bumpstop hit channel, solve big guess work and tedious task.

Also finding contact point via scatter will never be mm accurate and if there are slight differences in test run it will even show slightly different values for same settings......it is very very problamatic.

 

How many values are there?

You want to effectively come up with an x-y graph where your x axis is spring value (/clicks) and your y axis is bumpstop position. A simple equation like y = ax^2 + bx + c should get you close, where you choose appropriate values for a, b, and c.

Note that high school level algebra is very handy for this stuff. Remember those kids who complained "I'll never use this stuff!"? They were wrong.

Of course, you'll need to add ride height in there as well in reality.

Option 2: don't bother. Run against AI, or join/create fixed setup servers.

 

option 2....hell no, i want to know about setup to decent degree.

For corner springs there are 32 clicks/values for both Front and rear
and for 3rd spring there are 47 for both front and rear, iirc.

I can see how a line equation could help with this but the total travel is affected by both corner and 3rd
so there are 2 X.

Another logic i have is to form a equation such that whenever suspension forces becomes non linear it gives a value which i can assign to activates bumpstop gauge/alarm.

Edit: This might be wrong and stupid ...but i'm goona type anyway,
lets say if spring rate is 100N/mm, so could this mean all suspension forces (without touching bumpstop) are multiple of 100 ?
So if susp force is not divisible by 100 or it is not a multiple of 100, then return 1.... hence bumpstop is activated ?

 

We aren't actually drawing a graph, so you can have as many inputs as necessary. All you need to do is work out the relationships.

The reason you analyse the travel using a scatter plot is you can't rely on individual points being completely accurate.


The hypothetical I raised earlier was to point out that the actual use of this information isn't just to know whether you are on the bumpstops at a particular moment, but to be able to measure and analyse how much bump rubber compression you have. You really don't want to manually check what suspension position "bump stop contact" changed from 0 to 1 and work out how much particular (track) corners might be affected, or even have a constant value in the Maths that you adjust for each setup change. Hence the formula stuff...

 

ngl, I still haven't figured put how will i go on about finding the realtionship between stuff and use the formula you provided.
maybe something will "click " later in my brain, there are other things i also thought about.
But thank you for your help as always.

 

I think you are overcomplicating things a lot.
From what I've noticed, if the motion ratio of the car is 1, then ride height in setup - packer length = approx. ride height you hit bump rubber.
When the motion ratio isn't 1 the offset to bump rubber isn't that simple when you change ride height or packer, and you'd likely need the motion ratio to get an accurate calculation.

Normally you don't really use the corner bump rubber much on any car unless it's forced from setup or your last/best option to keep the car from grounding. If the car has a 3rd bump rubber you only have to use that to control grounding or get the ride height in whatever window you desire.

No fancy calculations or documentation is needed, just keep it simple

 

Truly it is over complicated. i understand your point using 3rd packers to stop bottoming but to accurately determine that i will need precise point of bump contact.
all this wouldn't be this complicated if only there was simple channel for bumpstop contact for both corner and 3rd .that is harder and bigger part of problem, after that i'm good to go, nothing to worry about really.

 

A scatter in Motec with suspension load and travel is enough to determine that. If you want to go further you can note the suspension position at the resting point and determine the amount of free travel before hitting the bump rubber.

I've worked with setups on many cars in the game and I've never had to do any advanced thinking regarding the bump rubber, so I'm confused over all this discussion on something so basic.

 

Right so i'm reviving this old thread for a bit.
Does anyone know how to calculate 3rd springs suspension force ?
there are channel for corner springs forces, i want same for 3rd springs.

 

Front/rear 3rd suspension pos is the avg of the corner springs, so I assume the forces are the same.

 

The problem i'm facing is i can't tell from scatter plot if 3rd spring is on bump or not. because for time being i'm using corner susp forces to scatter plot 3rd but at same time
i'm utilising corner packers so its hard to tell. it actually generates same scatter plot for both corners and 3rd spring at this point.
So if there is a way so i can get 3rd spring susp force only , then i can use that in scatter plot.

 

The 3rd springs are literally attached between the corner spring attachments. That's how it works.

You'll need to back off/remove the corner packers to try and isolate the 3rd packers. You may need to add a lot of 3rd packers so you can easily spot it (maybe very high, then medium, and gauge it from there).

 

Yea i did this exact method but in skid pad testing.
right now i'm satisfied with how rest of my setup is, except 3rd packers.
I thought maybe i can find out 3rd susp force only without corners involved somehow.

 

If you have all the spring and damper values of the 3rd suspension you should be able to create your own force channel, but that is really extra work for not really any beneficial gain. Although if motion ratio is anything but 1 and you don't have the exact value the force will be inaccurate.

To check when you are riding on the corner bump rubber you can look at a scatter with suspension roll angle vs lateral G.


This example picture shows no riding on the corner bump rubber, so the data is very linear. When you hit corner bump rubbers in high-speed corners the roll stiffness will increase and you will see non-linear behavior. It's very easy to spot if you compress the bump rubber. If the bump rubber is really stiff the roll angle can almost flat line.

 

I think i have been able to utilize rear corner bump to only touch at fast corer (pouhon at spa) using scatter plot of rear susp force to rear susp travel.
This is new method u have shown here, looks interesting i will try it out.
but before that i want to know how u find suspension roll ?
A method i know of is using some trigonometry
so for example at rear i use

atan ((RL wheel Y location - RR wheel Y location) / rear track width)

is this what u used to get suspension roll ?