Because of the disappointment about the handling of my 600 Abarth, I did another study. This time I looked into the suspension geometry.

I had a discussion with a friend of mine. He stated if you do not know, just do what Abarth did. Like I mentioned in the article about the spring rates: you cannot just buy parts and install them, without exactly knowing what that part will do for your particular setup. So I do not exactly agree with him.

Although this is not why I started this study. I did not like the way the car handled, so I looked into the suspension. In the last article, it was about spring rates and I think these can solve a big part of the understeer issues. Besides understeer, I had the feeling the steering was everything but responsive. From my basic knowledge about suspensions, I found that my double wishbone setup did not look right. Lowering the ride height initiates the problem, which results in an increase in body roll momentum.

Deflected angles of wishbone setup

The theory behind the determination of body roll momentum is not explained in this article, but can be found all over the internet. To at least understand this article you have to know that race engineers want their roll center beneath, but as close to the center of gravity as possible. A car tends to roll more, when the roll center moves further from CoG. The roll center can be determined graphically if you measure the rotational points of the suspension and track width. By looking at the angles, I implied the roll center was moved further from the CoG.

Graphical determination of roll center F1 car.

I measured all the rotational points of front suspension plus track width and made a graphical approach of the Roll center. Sadly, height of CoG is hard to measure, but can be estimated. I tend to locate it 35% of vehicle height.

Figure 1 shows the graphical approach as it is now. The roll center is just above the road, which is not that bad, but can be improved. Also keep in mind that the car is not lowered that much in comparison with some other Abarths, but I expected a worse scenario.

Figure 1
My Abarth. Not lowered that much in the front...

This is the time to return to the discussion with my friend. He introduced this worse scenario by just installing parts that Abarth installed in the early days. Not knowing it will result in a negative effect to the way the car will handle. I am talking about upper-arm camber correction. These will shift the upper arm upwards by +/- 1.5cm or even more and give the arm a more neutral position. Yes, this will correct camber, but is redundant when you have a fully adjustable independent suspension. The purpose of these camber corrections are intended for leafspring setups. Figure 2 shows what will happen if that rotational point is raised by just 1.5cm. The instantaneous center inverses and the roll center will be below the road and move further away from CoG. Roll momentum increases, which will give more stress to suspension parts. Imagine the anti-roll bar has to work harder now (i.e. you can hold a case of beer on the same height longer when it is close to your body, then when you hold it with a stretched arm parallel to the ground).

Figure 2
Camber correction adapters. These shift even more than 1.5cm.

This is just my theoretical approach and I hope someone is open to discussion on this topic and this particular part.

The question is how we can close the gap between the roll center and CoG, besides putting a shitload of weight underneath your car. Graphically it is easy; I lowered the outer joint of the lower arm by 4cm. As figure 3 shows, the roll center is moved above the road. This should be a big improvement.

Figure 3

Actually, this is done in many modern racing cars. Modern (performance) cars are well engineered and have their roll center close to CoG. In modern racing cars they lower the outer joint with a thick spacer to correct the roll center location, due to ride height adjustment.

BMW E46 race car with roll center correction at lower ball joint

Abarth made a piece to lower this joint, which are intended for lowering ride height on leafspring setups, also these are quite expensive. When I look at them, I do not believe they are very rigid and will introduce another joint. An alternative option would be some sort of aluminium spacer or weld extra tabs on them. The last thing is done by Fiat 850 enthousiasts. However, like the Abarth solution it was intended to keep the original leaf spring and lower the ride height.

To be continued...

Lower joint adapters (ABARTH), which lower the joint by 4 cm.
Welded tabs to Fiat 850 knuckles