Suspension Tech 101 - how does it work?

[gldgti]

This is intended as an educational tool. It is not a guide to upgrading your suspension, but information about how suspension actually works, and what actually goes on between the road and the chassis mounts - in terms of energy and response. It might all seem a bit unnecessary, but an appreciation for the true physical role of certain components in the system allows you to think about and diagnose what you percieve when driving your car to good advantage - you can make informed choices about modifications to the system rather than rely on the advertised results, and better understand whats going on between you and the road.

Whether you glean anything from my explanation is another thing, but I'll do my best

A note - this is in the realm of science and engineering. This is not the same as 'racecar engineering'.

This thread may be updated as I feel like adding more. If you see a typo or error, let me know.

[gldgti]

A lot of people have a lot of theories about suspension, and most people don't really understand what is actually going on. The basic physics are reasonably complicated to the layman, but its easily within the realms of an enthusasts understanding.

The overall system

A cars suspension exists for just one purpose - to convert vibrations from the undulating road surface into 2 things - heat energy, and vibrations of a different frequency.

The heat energy comes from kinetic energy that is converted into heat energy by deformation of the rubber tyre, and rapid movement of the oil contained in the shock absorber.

Why do we need to absorb this energy? Well, as the car drives along the road, and the road surface undulates, obviously without any suspension the car will need to follow the contour of the road exactly - not very practical. If we used only a spring system, with no damping (energy absorption) then we can certainly change the frequency of the vibrations transmitted to the vehicle - the trouble is that because of the vehicle mass (which is very large) and the added problem of vehicle inertia comressing/extending the springs with large changes in direction, the system can become both highly energised and uncontrollable.

Springs do not absorb any energy (in any practical sense atleast) - they only store and release it and various frequencies.

This means that our car suspension system, in order to work properly, needs to absorb energy - so we have tyres with both spring and damping characteristics (tyres absorb some energy when they deform - and they heat up when this happens) and we have oil filled dampers. Anyone who has ever played with an oil shock absorber will appreciate it takes energy to compress/extend it, but it doesnt give anything back! (Don't confuse this with a gas/oil shock absorber, which has some gas inside as well as oil - this is to change the way the damper reacts (usually in one direction only) once the gas heats up - kind of like having shock absorbers with 2 damping rates, on the go - neat huh?)

The effect of these 2 components is wonderful - the full magnitude of the bump/undulation is not transmitted to the vehicle chassis - instead, some of that kinetic energy is converted into heat, and dissipated to the atmosphere.

Of course - its not as simple as all that in practice -

What the hell is damping and where can I get some?

What is damping? We know we need it... but what is it? Well, its essentially a rate. Rate of what? Well, its a rate of energy required per unit of velocity to make a certain mass move. Note well - its related to velocity - but the velocity of what exacltly?

Well, the velocity of the mass.... what mass? Good question - there are 2 that we care about - the car itself, and the suspension mass, including the wheel and everything else that moves. (PS - the way its connected has a big effect, but that definately not within the realms of "101" level course).

Lets go back and look at what the suspension actually is for a second: Between the road and the chassis, we have:

Tyre
Rim
Bearing/hub/brakes/driveline/etc etc etc
Spring/damper
chassis

...or thereabouts, depending on the vehicle.

but, what you really have there is

Spring/damper/mass
mass
spring/damper/mass
mass

Thats all it is - 2 springs/dampers connected in serial with 2 adjacent masses. This is what gingerbeering folk call mass damper systems, and they can be anaylised by themselves and together - but the fact that they are always working together in our car suspension makes it a lot more sensible to analyse them together.

So, mathematically then, whats going on?

Well, a spring has a spring rate - an amount of energy to deform the spring a certain deflection.

A damper has a damping rate - an amount of energy to deform the damper at a certain velocity.

And the mass - well, it provides most of the inertia that forces the spring/damper to compress or extend - imagine a car that weight nothing at all - the springs wouldnt compress much over bumps now would they?

When you put a spring and a dmaper together, connected to a mass, what you get is a dynamic system, and this dynamic system can be analysed for its frequency response.

Holy crap you're thinking - now this is getting complicated - well yeah, but we'll keep going. Stay with me

What the HECK is frequency response?

Its what comes out the other end. It is literally the systems "response" to input, specifically in this case, the input is a vibration, or oscillation, with a certain frequency/magnitude. The requency is the "speed of oscillation" if you like. The magnitude is the deflection of each spring/damper coupling. This is probably a good time for an example:

When your car is at rest, it is static. There is no "input" to the system - this means that the systems response is (predicatably) zero.

What about when the car is driving on a completely, perfectly flat road, at a constant speed? Ofcourse, the input is still zero - we are only interested in something that causes the suspension to "deflect" from its neutral position.

So - what is the input? Well, that kind of depends on your perspective. Often in dynamic systems, we think of one "end" of the system (perhaps the chassis or the road) as being a point of reference, and working from there. Thats OK so long as you can work it so that you can calculate everything you need to know.

What is probably suprising is that when modelling car suspension, we want to look at it from the "road"



In this (pretty poor) picture, we have a diagram of the suspension represented as the 2 connected mass/spring/damper systems, with ground being the road.

In our case, the ground provides the input, and our deflections aremeasured relative to the neutral position (static). So, the deflection of the suspension strut is the change in distance between the wheel hub and the chassis, when the system is "exceited". The deflection of the tyre is the change in distance between the road surface and the wheel hub - as the tyre squashes due to load.

Inputs - what are they then?
Well, our system input is a result of the wheel following the contour of the road...... basically. The fact of the vehicle mass having some inertia means that as the tyre rolls over an undulating surface, the type and suspension undergoes some deflection. These deflections can be fast or slow, small and large, and this is where the frequency, and magnitude parts come in. Its safe to say the range of input freqencies and magnitudes possible for a car suspension is very large - frequencies vary from zero to hundreds or thousands of hertz (like driving over those ripply lines on the freeway, or very fast over a "washboard" rough road). Lower frequencies are those encountered by large dips or humps in the road, or even cornering. Magnitudes are clearly the result of the size of undulations - but also remember that the inertia o the masses in the system is what creates the deflections in the first place - likewise, the springs/dampers are the direct reason for the masses response to the original input!

It might be obvious, but the output of the system here is the nature of the movement of "mc" - the car mass. So, it should be obvious that if you change ANY of the parts of the system - the tyre, spring rate of the suspension or damping rate of the suspension, you will change the response of the system to a dynamic input.

But what effect does any given change have?

Tuning of a dynamic system

I wont make out that tuning a 2 spring 2 mass 2 damper system is simple - its not. Add to that the fact that there is one of these systems at each corner of the car, and it becomes more complicated. Add in things like anti-roll bars, etc, and it really becomes a task for a team of clever dudes with computer models of the system, working out what they want the system's response to be to certain inputs and tuning the system components (changing spring rates, damping rates and masses) to acheive the desired output.

How are the systems tuned in the first place? Well, the manufacturer decides how they would like the system to respond to "typcial" input - i.e. - they look at typical road sufaces likely to be encountered, typical driving speeds, and they modify the system to make suspension and the vehicle mass respond the way they want. But, its is ALWAYS a trade-off. There is always an engineering trade off when tuning a dynamics system, and suspension is no different. So, your standard mk1 golf suspension might seem soft, squishy and tall, but you know that its been engineered as a trade off between all kinds of scenarios - handling, wet grip, dry grip, bumpy road grip, comfort - to name but a few things!

You can change the system to suit your needs, but you will compromise some other area - some other response - at the expense of improving the response in another area.

[gldgti]

What makes a dynamic system tick?

So, how will it respond? Well for ANY given input, there are some things you can rely on -

1) Increasing spring rate will increase the speed of the system response.
This means that not only will "sprung" parts of the system (parts that arent ground) react faster to input, but the frequency of the response also increases. This is like saying that for a soft spring, the car may bounce "slowly" over some given bumps, wheras with stiffer springs, the same car will bounce "quickly" over the same bumps, at the same speed.

2) Increasing mass will decrease the speed of the system response.
Increasing mass has the opposite effect of increasing spring rate - that is, a higher mass slows down the response to an input - like if you take the same car, one empty and one full, and drive over the same road and at the same speed, the one with more mass will bounce more slowly over the bumps - if you watch other cars on the freeway you can ovserve this all the time.

3) Increasing damping rate will decrease the magnitude of the response.
This is very important - the damping does not change the frequency of the response of the system - only the magnitude.

OK, so what does all this rubbish mean?

Types of response of the system to input

Under damped system
Most of us have driven a car with worn out dampers - its "bouncy" (the output magnitude is large), and the suspension usually responds bady to bumps in the road (the wheels skip over bumps). This is because usually, the dampers have worn out, meaning that the damping rate has significantly decreased.

This scenario is called an Under Damped system. In this case, the system may continue to oscillate after a single oscillating input (the classic "bounce test"). This is because the spring rate and masses have stayed the same, but the damping is gone - now less energy is absorbed as the suspension deflects, and because of this the energy is instead trasmitted through to the car chassis (the primary mass). Furthermore, it upsets the reposnse of the suspension (the unsprung mass, which isnt really unsprung at all now is it?) which may be ocillating over bumpy roads instead of staying in contact with the road).

Critically damped systems and over damped systems
Most suspension systems are tuned to be almost critically damped to typical inputs. Critical damping is a special case, in which the response of the system to an input returns the system to zero without oscillating, but in the fastest time possible for the system components.

In essence, any more damping than "critically damped" is the system responding slowly. Over damped systems return to the neutral position without any oscillation at all.

The downside to over samped systems is that as the damping rate increases, more and more of the input is transferred directly to the output of the system - in the extreme case, the dampers are so "hard" that they effectively will not allow deflection (and hence will not absorb any energy) for inputs that have a high frequency.

If you have ever driven a car with sporty suspension on a very bumpy road you will likely have experienced the effects of this - Lots of vibrations are transmitted into the chassis of the car, and also, the wheels may bounce over the bumps in the road (and cause a loss of traction) because the suspension mass cannot oscillate quickly enough to stay in contact with the road.

[kaanage]

You should also take the opportunity to tell people here what the proper terminology for a couple of basic suspension components
Everyone here seems to talk about "sways" and "shocks" which are stupid Americanisms that have crept in over the past decade or so. The clued up tuners in the USA used to prefer the English terminology since it actually described the function of the components while their terminology was misleading.

IE there was a thread started this week where someone wanted to reduce the roll of their car and wondered if "sway" bars would have the right effect. If he knew them as anti-roll bars, then he would never have had any doubt.