The online racing simulator
Spring preload
(106 posts, started )
Spring preload
Hi

In a vain attempt to get more technical discussions going, I thought I'd open up this forum to the 'debate' on spring preload. Allow me to set the scene (yes, this applies to the F3 car, but for the most part you can ignore that).

Springs. Connected to the wheels via pushrods and bellcranks, so the wheel rate varies (lets say approximated 1.8:1 wheel movement:damper movement at full droop, to about 1.2:1 at full bump). We run the car approximately two-thirds between droop and bump, with the most travel being in bump. We are not worried about the wheel going into full droop.

A few people suggested our springs were too soft, and that we should go harder. Fair enough. We were running 275lb/in front, and 325lb/in rear. After working out wheel frequencies, I settled on 325 front and 350 rear. Keeps the nose off the track without going pointless hard.

However, this is where it gets confusing, and I'm of the opinion that most club racers and a lot of website writers don't know what they are talking about... Spring Preload. People have told me that I should run, say, 1 inch preload at the front and 0" preload at the rear. Their reasoning is that more preload effectively makes the spring stiffer, keeping the nose off the deck and improving the balance of the car.

I know, from very basic spring mechanics, that preloading a spring does NOT change the rate of the spring. If it takes 300lb of preload to compress a spring 1 inch (i.e. a 300lb/in spring) it will take another 300lb to compress it a further inch. It also does not, as some people have tried to tell me (and these people have been runnning single seaters for a while, and are winning!) mean that the first 300lb of load on the spring do not compress it.

So, as far as I can see they don't have a clue really. Raising the spring platform (i.e. increasing preload) will do the following:

Raise the rideheight
Move the bellcranks, and change where on the rising rate curve the car operates
Anything more than zero preload stops the springs rattling free at droops.

It does not:

Increase the rate of the spring (until perhaps just before the spring goes coil bound, but that's not likely)
Absorb the first x pounds of load on the spring

Yes?

Motorbikes often talk about preload - presumably this is a ride height change, as you do not have rising rate forks/swingarms? Or do people in the motorcycle world think it might do something else.

There is a chance I am wrong, which is why I'm opening this up for debate.

This is a website I was sent today - http://www.tftunedshox.com/techtips.htm which claims that the spring gets stiffer - is he an idiot, or is he correct)
#2 - ajp71
Quote from tristancliffe :
However, this is where it gets confusing, and I'm of the opinion that most club racers and a lot of website writers don't know what they are talking about... Spring Preload. People have told me that I should run, say, 1 inch preload at the front and 0" preload at the rear. Their reasoning is that more preload effectively makes the spring stiffer, keeping the nose off the deck and improving the balance of the car.

Ignore a lot of club racers' opinion, often these kind of bits of advice are given based on experience and passed on over time, alot of the time there's no logical checking in the thought process so sometimes people come out with complete bollocks, in the end of the day if you've done A-level physics you've probably got a better understanding than most people at a club meeting or IIRC in your case an engineering based degree you're certainly fairly over qualified. Unfortunately it can be hard to convince people otherwise regardless of the fact they're clearly totally out of their depth and don't understand what they're saying. We have quite a few people like that on this forum, often who have no actual experience or understanding of real life motorsport who trip up on pretty basic things trying to sound clever.

I could attempt to write you an answer but at least I recognize the fact I'm out of my depth and won't answer without confidence in what I write. Your as capable as anyone of reading up on such a topic and it's probably the best way of getting a proper understanding rather than just a random answer, which might be right, you'll certainly get lots of answers to try, ask 5 different racers what you should set your dampers to and you'll get 5 random numbers ask five suspension specialists and you'll get a couple of different answers with logic and reason behind them
#3 - w126
Real suspensions springs may be progressive (i.e. have slightly more lb/in when being compressed more) instead of being totally linear.
Let's ignore progressively wound springs, as single seaters don't use them.

Thanks Alex. I admit I have already done some reading up on the matter, but for every article that agrees with me I find 50 that agree with the club racers. I too believe it's one of those things that has passed down from winner to wannabe, and become fact rather than just something to think about yourself. I've seen the same thing in Top Fuel drag racers, F3000 cars, and lots of tintops - people do things because 'that is how it's always been done' rather than realising it might be either plain wrong or very inefficient. Not been close enough to professional racing to know if they suffer from the same misinformation passing around, but I'd suspect they do to an extent.

People have tried to tell me my front wing is designed as a ground effect wing because it's large aspect ratio. Whilst I'm sure the wing does operate slightly in ground effect, I bet that's not the primary reason for the large aspect ratios. In fact, I don't really know why a lot of cars have large aspect ratios - the most efficient wing is long and slender (see gliders or passenger planes). Maybe it's because length is constrained by the regs, so they go deeper looking for downforce.

Don't get me started on the erroneous beliefs about gearing either...

I just don't understand why people don't think. They plow loads of money into a racing car, either in purchase costs or running costs, then just accept at face value what their competitors tell them (even if the competitor isn't telling them falsehoods on purpose). They don't seem, on the whole, to be a stupid bunch of people, yet they seem to prefer not thinking to thinking.
from my experience with rc cars i do know that it does in fact make the springs feel stiffer despite not using progressive springs

im currently running into the same physical problems you ran into with hookes at explaining it but experience of playing around with rc dampers at various preload settings does confirm that theres some effect
#6 - ajp71
Quote from tristancliffe :Not been close enough to professional racing to know if they suffer from the same misinformation passing around, but I'd suspect they do to an extent.

I have no doubt they do certainly among the semi-professional series I've experienced a bit of it's not a lot better than club racing.

Quote :
People have tried to tell me my front wing is designed as a ground effect wing because it's large aspect ratio.

The front wing on our car used to be enormous, so much it was always on the lowest setting and we nicknamed it the platypus. People could come up with a million reasons why it needed such a massive wing that clearly produced far too much downforce and drag for the car and another million reasons why fitting smaller wings, like all the fast cars used would upset the air flow and make it randomly crash or some bollocks like that. Guess what when we fitted a new less deep and silly looking wing it was faster in a straight line, faster overall, didn't have to run full rear wing to compensate and didn't start randomly crashing

Quote :
I just don't understand why people don't think. They plow loads of money into a racing car, either in purchase costs or running costs, then just accept at face value what their competitors tell them (even if the competitor isn't telling them falsehoods on purpose). They don't seem, on the whole, to be a stupid bunch of people, yet they seem to prefer not thinking to thinking.

Yeah lots of things like the belief that running asymmetric camber will result in a car that will randomly crash when it feels like it :doh:
@shotglass (Alex's post wasn't there when I started): You mean you think springs do get stiffer with preload? That's just not the case.

Take a 100lb/in spring. If you put 100lb on it, how much does it compress? 1 inch. If you put 200lb on it, how much does it compress? 2 inches. That is why it's called a 100lb/in spring, not a 100 - 200lb/in spring, with a nice little graph explaining the displacment/rate curve.

All preloading is, as it implies in the name, is putting an initial load on the spring, so 1 inch of preload is the same as putting an extra 100lb on it. Adding the weight of the car (lets same it's 300lb per spring) make it compress a further 3 inches, not 2.5, not 2.0, not 1.5. The first 100lb of the 300lb load isn't 'absorbed' by the preload of 100lb, it's additive.

Sorry to sound assertive, but I want to understand this. Feel free to post why I'm wrong in as patronising a way as you please
#8 - JTbo
I have heard that by setting preload you can change ride height and that is about it. Of course position of suspension affects roll centre position etc. but in practise it would be just ride height and rest position of suspension that you are changing with preload.

Don't know from any better so I have trusted that what I have been told.
Quote from tristancliffe :the most efficient wing is long and slender (see gliders or passenger planes).

I believe you´re wrong on that one. From what I know about wings their performance mostly depends on two things
1) its profile (ie. 2D space; regardless of its length)
2) inability for the air to move laterally across the wing (think of it as a 3D defect)

The problems in 3D space are mainly caused by the air beneath the wing trying to fill the void above it by going around the tip of the wing, thus causing the eddy at the tip.

The ways employed to get rid of the problems appearing in 3D space are various.
1) endplates
2) winglets
3) long wings
All three of those effectively crunch the 3D aiflow down to a 2D one either by limiting the airs ability to get on top of the wing (1 and 2) or by moving the tip out to infinity.

The end result of all this is, that the aspect ratio of a wing only matters when you havent got anything bolted to the tip of the wing, that the only thing in effect is its profile and, that the only point of building a long wing is to get more wing profiles next to each other.

Quote from tristancliffe :You mean you think springs do get stiffer with preload? That's just not the case.

I don´t think it does, experience from compressing RC springs with and without preload between my thumb and index finger tells me it does.

I think what we can easily agree on is that a spring works according to Hooke´s law which is:

F = k*x

For simplicity´s sake we assume that the spring is fully extended when the damper is fully extended, ie. the springs length matches that of the damper or in other words the spring wont kill you if the gear used to put it into the damper fails since you dont need any.

So if we were to compress the damper, discounting that it dampens, the force needed to compess the spring by x directly follows from Hooke´s with x=0 being a fully extended damper.

If you add a preload of x_0 the force the sping exerts is, with x still being the _dampers_ change in length from full extension

F = k * (x + x_0)

Thus the force needed to compress the damper by x_1 which previously was:
F = k * x_1
has now increased to
F = k * (x_0 + x_1)
or in other words you have added a bias term of k*x_0 to the force.
Quote from tristancliffe :People have tried to tell me my front wing is designed as a ground effect wing because it's large aspect ratio. Whilst I'm sure the wing does operate slightly in ground effect, I bet that's not the primary reason for the large aspect ratios. In fact, I don't really know why a lot of cars have large aspect ratios - the most efficient wing is long and slender (see gliders or passenger planes). Maybe it's because length is constrained by the regs, so they go deeper looking for downforce.

I think you've got your definition of aspect ratio wrong. Aspect ratio is calculated as (span^2)/wing area, so a long, thin wing (like you'd find on a glider) has a very high aspect ratio. The wings you describe for your F3 sound like low aspect ratio wings, having short span and long chord.

Higher aspect ratio wings are more efficient because the induced drag is reduced. Induced drag is caused by the wing-tip vortices of any finite-span wing creating a downwash region between the wingtips, hence changing direction of the lift and drag vectors. A component of the lift vector now acts to cause drag and a component of the drag vector acts against the lift. The wing-tip vortices of wings close to the ground are disrupted by the presence of the ground and the induced drag effect is smaller (search for the Ekranoplan to see this in action!).

What this means is that, for wings close to the ground, the effect of a high aspect ratio is already provided by the ground! Of course, having a high-aspect ratio wing will help further, but it's not so important as providing maximum lift.
Another nice feature of low aspect ratio wings is that their stall behaviour is much more gradual than a high-aspect ratio wing.

Getting back to topic...I think you're right about suspension preload settings. Most of the sites I've found dealing with bike suspension suggest that preload is used to change the ride height of the bike and should be adjusted with driver weight, etc... If a spring is linear then the force required to compress it won't change with compression.
The only effects of changing preload should be to change the ride-height of the vehicle and to alter the static position of the bellcranks, as you've already noted.
Can anyone draw a quick diagram of how the spring is positioned/used in a suspension? Right now the only thing I can imagine preload doing is increasing the ride height.
Quote from Shotglass :I think what we can easily agree on is that a spring works according to Hooke´s law which is:

F = k*x

For simplicity´s sake we assume that the spring is fully extended when the damper is fully extended, ie. the springs length matches that of the damper or in other words the spring wont kill you if the gear used to put it into the damper fails since you dont need any.

So if we were to compress the damper, discounting that it dampens, the force needed to compess the spring by x directly follows from Hooke´s with x=0 being a fully extended damper.

If you add a preload of x_0 the force the sping exerts is, with x still being the _dampers_ change in length from full extension

F = k * (x + x_0)

Thus the force needed to compress the damper by x_1 which previously was:
F = k * x_1
has now increased to
F = k * (x_0 + x_1)
or in other words you have added a bias term of k*x_0 to the force.

The extra force of k*x_0 is the preload! That was added when you tightened down the preload collar and has nothing to do with the force needed to compress the spring further.
F = k * (x_0 + x_1) is the force needed to compress the spring by the preload x_0 and the extra compression x_1. However, the x_0 distance has been added in the preloading and you only need to compress it an extra x_1, hence the force required is still k*x_1.
Android - Increasing? I always thought it would have decreased.
Quote from StewartFisher :The extra force of k*x_0 is the preload! That was added when you tightened down the preload collar and has nothing to do with the force needed to compress the spring further.
F = k * (x_0 + x_1) is the force needed to compress the spring by the preload x_0 and the extra compression x_1. However, the x_0 distance has been added in the preloading and you only need to compress it an extra x_1, hence the force required is still k*x_1.

GAH
that force is still in the system currently trying to lengthen the damper (which it cant) and you first have to equalize that to change the springs length to any degree

forces dont just fade away into thin air
Quote from Shotglass :I don´t think it does, experience from compressing RC springs with and without preload between my thumb and index finger tells me it does.

If I understand you correctly, don't you limit the spring + preload maximum length to a fixed value? Say you have a 3cm spring in a 3cm "casing". Then you add 1 cm preload, with the total length still being 3cm (the spring got compressed to 2cm). If the spring had a stiffness of 1kg/cm, you'd now have to exert more than 1 kg pressure to actually compress it, anything less would simply leave it fully extended at its 2cm. The thing is though, I think, that the spring in a car is always compressed to some extent and the car never really rests on top of a fully extended spring (that is strong enough to push the car to the limit of its "casing" or whatever limits the spring length (shock absorber length?)). I might be talking complete and utter bullshit, though
Quote from Bob Smith :Android - Increasing? I always thought it would have decreased.

Don't worry, it's probably me just having no idea how an actual suspension works.
Quote from Shotglass :forces dont just fade away into thin air

How exactly, from a mechanical viewpoint, is pre-load achieved? I can certainly think of at least one way of making pre-load disappear when spring deflection is larger than pre-load.

Quote from AndroidXP :Don't worry, it's probably me just having no idea how an actual suspension works.

That's fine, neither do I really.
Quote from AndroidXP :The thing is though, I think, that the spring in a car is always compressed to some extent and the car never really rests on top of a fully extended spring (that is strong enough to push the car to the limit of its "casing" or whatever limits the spring length (shock absorber length?)). I might be talking complete and utter bullshit, though

this is of course correct and as such adding preload will undoubtedly have an effect on ride height if you dont do anything to counter that (shortening the pushrods or mounting the damper differently)
this was mainly to just isolate the effects of preloading a spring

btw in fact on a real car the spring is pretty much always preloaded even with the dampers fully extended (thus the comment on the spring killing you if the gear used to mount it on the damper fails)

Quote from Bob Smith :How exactly, from a mechanical viewpoint, is pre-load achieved? I can certainly think of at least one way of making pre-load disappear when spring deflection is larger than pre-load.

http://sugi5476.blog.ocn.ne.jp ... _blog/images/damper_2.jpg
do you see the thread on the damper casing ? the knurled blue thing at the top of the spring is used to preload the spring by scewing it further down
Quote from Shotglass :http://sugi5476.blog.ocn.ne.jp ... _blog/images/damper_2.jpg
do you see the thread on the damper casing ? the knurled blue thing at the top of the spring is used to preload the spring by scewing it further down

Now find me an animation of it all working (i.e. showing the spring compressing). Or just answer this: basically, is the "knurly blue thing" compressing with the spring or not?
Quote from Bob Smith :Now find me an animation of it all working (i.e. showing the spring compressing). Or just answer this: basically, is the "knurly blue thing" compressing with the spring or not?

yes it is
its fixed to the damper casing by the thread on it
Quote from Shotglass :GAH
that force is still in the system currently trying to lengthen the damper (which it cant) and you first have to equalize that to change the springs length to any degree

forces dont just fade away into thin air

Let's say we have a spring which is 20 cm long and has a spring rate of 50 N/cm. It is sitting on a hard, flat surface. We also have 2 weights, each weighing 100 N (10 kg). Now, you'd agree that by Hooke's Law, placing all three weights on the top of the spring should compress it by (2*100)/50 = 4 cm, leaving the spring 16 cm long?

Let's add the weights one at a time. Weight number one will compress the spring by 100/50 = 2 cm.
Now let's add the second weight. By your logic, the spring shouldn't move, since there must be an extra 100 N of 'preload' in the spring because of the first weight. The extra 100 N weight will simply balance this 'preload' and the spring won't change in length.

But we agreed before that the spring should only be 16 cm long...
Quote from StewartFisher :Let's say we have a spring which is 20 cm long and has a spring rate of 50 N/cm. It is sitting on a hard, flat surface. We also have 2 weights, each weighing 100 N (10 kg). Now, you'd agree that by Hooke's Law, placing all three weights on the top of the spring should compress it by (2*100)/50 = 4 cm, leaving the spring 16 cm long?

Let's add the weights one at a time. Weight number one will compress the spring by 100/50 = 2 cm.
Now let's add the second weight. By your logic, the spring shouldn't move, since there must be an extra 100 N of 'preload' in the spring because of the first weight. The extra 100 N weight will simply balance this 'preload' and the spring won't change in length.

But we agreed before that the spring should only be 16 cm long...

GAH again

preload doesnt add weight ... its a weightless force on the spring which is used to deform the damper and its casing in an elastic manner (the damper being a streched spring of its own)
you then have to push the damper back into its original form before you are able to compress the spring
Would this be wrong:

(I'm going to read the last 10 posts now )
Attached images
susp_preload.jpg
That makes it look like applying the weight of the vehicle to the springs acts in the same weigh as pre-load, which I'm pretty sure would be incorrect.
Quote from Shotglass :GAH again

preload doesnt add weight ... its a weightless force on the spring...

How does the spring 'know' whether the force is weight or preload? The spring simply changes its length according to the applied forces, whether they're weight or preload.
Quote :...which is used to deform the damper and its casing in an elastic manner (the damper being a streched spring of its own)
you then have to push the damper back into its original form before you are able to compress the spring

I don't understand. How does the damper act as a spring?
Quote from Hyperactive :Would this be wrong

yes since weight should only move the equilibrial point to the right but not up

Quote from StewartFisher :How does the spring 'know' whether the force is weight or preload? The spring simply changes its length according to the applied forces, whether they're weight or preload.

the spring doesnt but the bits attached to the spring sure as hell feel the difference

Quote :I don't understand. How does the damper act as a spring?

because the damper at full extension is essentially a metal rod

Spring preload
(106 posts, started )
FGED GREDG RDFGDR GSFDG