I had the pleasure of speaking with Cary Eisenlohr at length regarding suspensions and bump steer in particular. Cary, in case you don’t know, is considered very wise in this area by many.

Bump steer, for those that don’t know, is the unwanted, and often undesirable, steering effect that results from toe changes during suspension compression and extension. I have focused on this aspect of the suspension effects since it seems with all of the lowering done by many, it is simply not addressed. I had addressed it when I lowered my 280Z by shimming the steering rack. This cannot be done with the 996 so I wanted to see what the real effects were and what solutions, if any, were available.

The first order of business was to see what the geometry looked like. To do this, I needed to measure the control arms, the steering tie rods, the suspension pickup points, etc. I then recreated a front and top view on CAD. In addition, realizing I lowered the car 40mm (1.57 inches), I extrapolated the stock geometry by simply raising the chassis points by that amount and adjusting the other parameters as necessary.

A little background before I continue. My big concern was that the control arm (CA) and the steering tie rod (TR) were angled in such a manner, following the lowering, that I would incur odd response characteristics from the steering when the suspension was stressed in a corner. In a perfect world, the CA and TR would have the same hinge points and therefore the arc scribed by each would be identical and concentric. Well, we are not living in a perfect world and the CA and TR have different lengths (13.5 and 16 inches respectively) as well as different hinge points. This, by its very nature, equates to some bump steer.

The results………….

On my lowered car, the CA and TR angle UPWARD from the chassis to the wheel by approximately 6.1 and 5.1 degrees respectively.

In contrast, for a stock USA turbo, the CA and TR angle DOWNWARD from the chassis to the wheel by approximately 0.87 and 0.72 degrees respectively.

What does this mean???
Well, as a stock suspension is compressed, because of the portion of the scribed arc (circle) the CA and TR are operating on, the inward/outward movement of the wheel is minimal. Upon first compression of the suspension, the wheel moves imperceptibly outward and then, upon further compression, as the CA and TR pass the horizontal plane, the wheel moves imperceptibly back in. In the lowered car, all movement is inward.

What are the real numbers???
After recreating the front and top geometry on CAD, I went through the exercise of seeing what the toe changes were for the stock vs. lowered scenarios.

Due to the simplistic front suspension, the change in toe is simply due to the differential inward/outward action of the CA and TR translated to the horizontal plane .
In real terms, for 2 inches of suspension travel (alot in my opinion):

• The stock car undergoes -0.25 degrees of individual wheel toe change (-0.50 combined).
• The lowered car undergoes -0.71 degrees of individual wheel toe change (-1.42 combined).

Keep in mind, these numbers are rudimentary but they are accurate enough to show a trend. An interesting trend to say the least.

Ideally, when a car is lowered, both the CA and TR pick-up points at the wheel should be lowered as well. This will allow them to operate as designed from the factory such that they are horizontal in the middle of their travel range. I am not aware of a product that allows the CA pick-up point at the wheel to be lowered but I am aware of a product that facilitates this at the TR pick-up point at the steering knuckle at the wheel. Steve Weiner at Rennsport sent me the following pictures pertaining to conical spacers that lower the TR pick-up points (the pics are not from a 996TT). They are actually made by Cary as I understand it. They seem like a very suitable and solid product.






Although I realize the CA and TR travel arcs should be concentric, thus necessitating the lowering of BOTH pick-up points, I nonetheless wonder what effects I would experience if I use these spacers to solve the TR upward angle issue only.

I would appreciate any helpful comments or discussion.

 

How does the factory deal with these geometry changes with their X73 kit , which lowers the car almost as much as yours ?
Do they just figure you have to live with the bump steer on public roads because it's not as much of an issue on smooth race tracks which is what the X73 is designed for ?

 

MKW,
Good question! I wonder myself. As I understand it, the X73 kit is available for european spec'd cars. We have used our resources to import it to the US. Could the euro cars spec'd with the X73 kit have a different wheel carrier? If so, it could conceivably have lower pick-up points.

I have noticed, with the stiff shock setting, on smooth pavement, as you have stated, the suspension works fine. I only notice some oddity when the car is unsettled in a corner.

So, maybe you are right. I would be very interested to know if the wheel carrier is different on the euro cars spec'd with the X73 kit. Gert? Anyone?

 

While it is an important factor that the CA and TR are as close to horizontal as possible, there's also the factor of keeping their pickup points as close to the same plane as possible. To deviate too far would exacerbate bump steer.

Imagine for a second if instead of lowering the pickup point of the TR just an inch, we instead lower it to infinity. You'd result in massive (infinite) bump steer. This is only meant to help visualize the impact of lowering only one of the pickup points instead of both. Of course it is possible that in lowering one pickup point, you are in fact closing the gap of getting them in the same plane.

Imagine a flat surface whose plane is identified the following four points:

CA pickup point on knuckle (CA-1)
point along axis of CA at chasis that is perpendicular to CA1
TR pickup point on knuckle
steering rack joint (the other end of TR)

In plane simple english, as the suspension goes through its motion, the more twisted this plane gets, the more bump steer you will get.

Sometimes, manufacturers will use this to engineer deemed desired results, for instance, toe-in when braking for stability.

So, the million dollar question is... will lowering the TR pickup result in a more or less twisted plane?

To fully realize whether this lowering mod will help with bump steer, we'd have to examine the stock suspension geometry (starting point of calculation) before drawing conclusions.

Saying it in another way, in my humble opinion, it's not possible to draw a general conclusion that lowering the TR pickup will necessarily solve bumpsteer problems in a lowered car without examining the rest of the suspension.

I know this doesn't directly answer your question, but answer is simply more involved than knowing that the TR pickup has been lowered. Hope this "helped"...

 

Stephen,
I fully agree with everything you said. To expound on your comments in a slightly different manner, consider this.......
The CA and TR outer pickup points are in the same VERTICAL plane (plane parallel to the rotor and not to be confused with the plane you described in your post which I understand). The inner pickup points of the CA and TR are in different plane simply due to the TR being 2.5 inches longer than the CA. The TR is in front of the CA (fore/aft). They are roughly parallel to each other. Therefore, the TR will ALWAYS scribe an arc of greater radius than the CA. When they both operate near the hozontal attitude, the radii difference shouldn't have much of an impact. When they operate above the horizontal attitude, as in a lowered car, their arc differences will pull the wheel in different amounts for a given suspension translation. The CA will ALWAYS pull the wheel in farther due to its shorter length and smaller radius arc. Remember the TR is in front of the CA. So, with suspension translation, the wheels will always toe out under compression of the suspension in a lowered car with upward sloping TR's and CA's!!!! REVELATION!!! UGH!!!!

So let's ask these two questions:

1. Is there a component (spacer) available that will lower the CA pick-up point at the steering knuckle?
2. Is my quest relegated to trial and error with the conical spacers shown in the photos above?

 

Ken...I do not think you can automatically assume that a small amount of toe-out under compression (if indeed it is a small amount) is necessarily bad. If we presume that one sets the car up to have a slight (very slight) amount of static toe-in, then the progression to slight (there's that condition again!) tow-out under cornering induced roll may just dial out understeer...sort of like ackerman. Now...if we are talking about significant amounts of bump-steer (two change with roll) this f's things all up.

Thoughts?

 

WTF: I posted a reply to this post and now it is gone????

 

J,
Unfortunately, I have to assume it is bad. I have had conversations with TonyNJ and he is experiencing the same effects as best as we can both verbally describe it to each other. We intend to swap rides in the near future to see if we are in fact experiencing the same effect. I find it very difficult to understand that we are the only ones experiencing handling oddities at the limits. Tony currently has H&R springs only whereas I have the PSS-9 coilovers. Our drops are about the same for all practical purposes. I am thinking either people are simply not driving hard enough to experience these effects, they don't know what a great handling car is supposed to feel like or they are just keeping to themselves.
In any case, we are both determined to get to the bottom of this.

I am not talking about oversteer or understeer. I am also not talking about a slight static toe in. As I pointed out, the total toe change, by my rudimentary calculation amounts to -1.42 degrees combined. In my opinion, that is alot!

I really feel the solution is some sort of spacer installed between the control arm outer end and the strut ball joint in conjunction with the conical steering tie rod spacers shown above.

Does anyone know if there are control arm spacers available???