Suspension Geometry
Including Lowering, Roll Centres, Bump & Roll Steer, Camber & Caster
There is no doubt in the world of Aftermarket Tuning and Club Motorsport, that suspension geometry is one of the most overlooked factors. All too often you will see a car that has had literally tens of thousands of pounds spent on a high-power engine and braking set up, only to find that the suspension received a fraction of the budget and is often seen as an inconvenience. Many seem to view the fitting of suspension as merely a way to get the arch to tyre gap as small as possible. “Slamming” or “Dumping” their “Ride” is the primary objective and many consider the lower the better. If you know what you are looking for, you will see the difference between a track car and a show car.
Firstly lets look at the advantages lowering will always give us:
-
Lower centre of gravity
-
Reduction in aerodynamic lift (assuming rake is maintained front to rear)
-
Higher top speed
As we already found in the weight transfer section that corner speed is a function of track width, weight and CG Height then surely it stands to reason that the more you lower the car the faster it will be able to round the corner?
If only it were so simple!
When we first considered the weight transfer that the vehicle experiences around the corner a very simplistic approach was taken assuming a fixed level of centripetal force. The proportion of weight transfer the front and rear tyres experienced was assumed to vary depending on the static weight split of the car. This was then developed when we discussed roll stiffness which demonstrated how even with unfavourable weight distribution outright G-Force levels could match a vehicle with “perfect “ 50/50 distribution.
Unfortunately this simplistic approach does not take account of complicated dynamics of a vehicle rounding a corner. What we need to take into account is the vehicles roll centre – both front and rear which together make the roll couple.
Discussion of roll centres in depth would take a long time and there are many good texts available should you wish to research further. This discussion will identify the basics so that they can be considered in your project.
The Geometric Roll Centre location is what we will discuss here (if you are finding this bit interesting also try Google searching for “Force Based Roll Centre” which is the latest development of this which takes into consideration the dynamics of the roll centre).
The Fig 1. below shows a MacPherson Strut Geometric roll centre:
Fig 1.
Weight transfer is affected by the distance between the CG Height and the roll centre. We need to recognise that not all the weight transfer goes via the springs, dampers and anti-roll bars. A certain amount of the weight transfer occurs as the forces generated by the cornering feed directly through the suspension links to the tyres.
Ignoring unsprung weight transfer to simplify the discussion, weight can move:
via the roll centres - In this instance weight transfer is separate for front and rear. It can be varied simply by raising or lowering the roll centre relative to the ground. So a ride height adjustment to your race car, or a roll centre geometry change is a very valid tuning device.
or
via the sprung mass - The chassis is a rigid(ish!) structure which rolls around an axis between the front and rear roll centres. Therefore the sprung mass weight transfer is based on a mass whose centre is the centre of gravity of the entire sprung mass, the mean roll centres and mean track. This weight transfer is resisted by the springs, anti-roll bars and dampers.
So what does this mean in practice? High roll centres leave less weight to be transferred via the springs and vice versa. We want the springs and dampers and anti-roll bars to work so generally speaking we want the roll-centre above ground level.
-
A roll centre above ground decreases roll, increases jacking *, initially loads the outside tire and unloads the inside tire, improves turn-in and increases ride height.
-
A roll centre under the ground increases roll, creates anti-jacking, initially loads the inside tyre and unloads the outside tyre, diminishes turn-in response and lowers ride height.
* Jacking forces are the sum of the vertical force components experienced by the suspension links. The resultant force acts to lift the sprung mass if the roll centre is above ground, or compress it it is below ground. Generally, the higher the roll centre, the more jacking force is experienced.
However MacPherson strut geometry has a particular problem. If we take the diagram and move the wishbone position to mimic lowering it becomes quite clear the the roll centre moves a long way and is now below ground level. See Fig 2. Below
Fig 2.
Not only is it below ground level but it will now move around much more markedly from left to right whilst cornering. This usually results in a vehicle that is unpredictable, rolls a lot and generally understeers (assumption we are dealing with front axle only).
Now these Diagrams actually show an extreme amount of steering axis inclination (SAI- The angle of the strut viewed from the front). Most modern MacPherson strut cars do not have such extreme levels of SAI so in actual fact are even more sensitive to excessive lowering. You can try this yourself (we could do it but it won't fit on the screen) but you will need a LARGE piece of paper. Repeat the drawing but if you angle the strut more upright you will see the line that current intersects will make a virtual point much further out and that lowering can bring the roll centre way below ground.
The bottom line on MacPherson strut though is that it is unlikely you will have the roll centre in the right place if the arm is pointing upwards at the outboard end. If you cannot get the arm in this position you may be able to lower the balljoint pivot point or increase SAI in order to get the roll centre under control.
What about Double Wishbone?
There are many flavours of double wishbone suspension, some good and some bad. Just because a car has double wishbone suspension doesn't mean it's going to be good round corners. Likewise even a car with a good design of double wishbone suspension won't necessarily be faster than a well-designed MacPherson strut set up. The beauty of double wishbone suspension is a much more consistent location of roll centre, less SAI required (excessive SAI is needed to prevent excessive scrub radius on MacPherson Strut), Better ride, Better Steering response as steering axis is separate to the vertical motion (see latest Renault and Ford MacPherson Strut designs which feature this benefit.
It doesn't mean that necessarily “ slamming” suspension is the answer if you have double wishbone and it is still necessary to understand where the roll centre location is. However the designer of a double wishbone suspension is able to control within a limited range of motion both the roll centre and the camber angle.
Bump & Roll Steer
Although Bump & Roll Steer are caused differently, they are modified in the same manner so I will cover them both together. Bump Steer is more of a factor on a road and Roll Steer more of a factor on a circuit, but both are undesirable.
When the car rounds a corner and sufficient weight transfer occurs to compress the spring we will on the outside corner will move upwards as the suspension takes the load. The wheel on the opposite side will of course move downwards as it is unloaded. In an ideal world (unless the designer has a specific reason for doing so) the wheel will maintain the same amount of steering angle the driver has applied as it moves up and down.
Unfortunately in reality this is often not the case. A lot of modern cars seem to have a bump steer curve that conspires to put you into the nearest tyre wall on a circuit.
We have seen as much as 8mm of toe change on the Mk5/6 Golf Platform (hint you need the Whiteline Bumpsteer kit for this car). This meant the outside wheel toed out under load, and the inside wheel toed in. Now if you consider the fact that Ackerman geometry requires that we turn the inside wheel at a greater angle it is easy to see why this particular car needs attention.
So how do you measure Bump-Steer?
Try a Google image search for “diy bump steer gauge” and you will see what kind of construction is required. I made mine from wood and a simple £20 dial gauge.
Once you have your gauge all you need is a pen and paper so that you can measure the change in toe through compression and extension. N.B You will need to remove the front spring to make the compression measurements possible.
Establish where the suspension sits when the car is static on level ground. Then from pictures of the car on circuit or using judgement assess the maximum compression point and the maximum extension point. Remember this is a key area where excessive lowering will cause problems. Also remember that severe bumpsteer on full droop is the lest problematic so if your changes “move” the bumpsteer anywhere full droop is the best place.
To achieve zero bump the front end must be designed correctly. The tie rod must travel on the same arc as the suspension when the car goes through travel. Simply matching lengths and arcs to prevent any unwanted steering of the front wheels. In reality Zero Bump is unlikely but if you get 0.25 mm or less toe change per 25mm of vertical movement the driver will be happy!
Solutions to Bump-steer
It depends on the amount of the course the set up of the suspension. Generally speaking if you can move the steering rack you should be able to reduce the bump steer to acceptable levels. After measurements have been taken you will know what you are looking to do.
Now on most cars moving the rack isn't easy so you may elect to try and make the changes at the rod end itself. You can either invert the rod and (may need to drill the hub) that might work wonders with large ride height drops. More than likely thought a specialist “rose” type spherical bearing with spacers is what is required.
Camber & Caster
Camber angle is one that most car enthusiasts are familiar with, as we often see lots of negative camber on racing cars. Changing camber angle is sometimes easy and sometimes quite difficult but getting the correct amount depends on a number of factors not least the tyre choice. E.g Crossply/ Bias Ply tyres such as Avon ACB10 require very little negative due to the additional stiffness in the tyre through the sidewall. Karts are another example where due to the tyre choice very little negative camber is required.
Radial construction is far more common in the majority of modern tin top and single seaters though. The correct amount of camber must be chosen to suit the tyre temperatures.
Caster angle is familiar to us as the angle that on most cars enables the steering wheel to rapidly self centre enabling us to unwind lock from a corner by letting the wheel pass-through your hands. Another example of caster would be on the bicycle that enables you to ride no handed. Eliminate the caster in these examples and both vehicles would be very difficult to master.
Caster causes a jacking effect in the chassis as one wheel will move up and one wheel will move down. It will load the inside front and the outside rear around a turn. It also creates dynamic negative / positive camber. As you round a corner caster will increase the negative camber on the outside front wheel and increase positive camber on the inside front wheel. This is a very useful attribute for a road car where you do not want to run too much static negative camber in the interests of tyre life.
Caster has another effect in improving traction. Next time you go shopping see how much more easily a shopping trolley wheel wants to turn one way more than the other! Modern front wheel drive cars, use increasing levels of caster to enable them to put very high levels of horsepower and torque through the driven wheels successfully.
Increased positive Caster will also make the vehicle more reluctant to turn and also increase the self centring effect. However it should give more mid-corner grip and also better traction. Some people say you can never have enough, being realistic more than 10° is not going to win any prizes as after that it is a law of diminishing returns due to some of it's less desirable effects. Older FWD hatches may have a little 2-3° Caster so there is lots of room for modification!
