_________________________________________________________________________________

Rack And Pinion Installation

Rack and Pinion Installation and Front End Geometry

1. Lubrication:  Two grease fittings are provided so that at least one can be reached in the car. The rack and pinion is already packed with a very heavy duty grease. We strongly recommend that you lube the rack weekly during the racing season. Greases sold at parts stores for automotive use are generally not heavy enough for this purpose, but are much better than nothing. Remember: frequent and generous lubrication of the moving parts is the single most important thing you can do to ensure maximum steering gear performance. 

2a. Mounting type G racks:

When bolting a type G unit into the car, use screws long enough to obtain at least one full inch of thread engagement (less than this may result in ripping the threads out of the magnesium) and coat the threads with an anti-seize lubricant. Do not use thread locking adhesives on cap screws that are to be tightened directly into the rack housing. To lock cap screws against backing out, use split-ring lock washers. It is also essential that the mounting bracket be reasonably flat and that it not interfere with the housing. On some older chassis the mounting bracket may be too wide to clear the rack housing, in which case you can either use shim washers under the housing or modify the bracket for proper clearance. Make sure to have at least a quarter inch of daylight between the rack and your crank pulley (engines tend to shift during impacts, and if the edge of a large pulley or harmonic balancer contacts the rack it can bend the crankshaft snout). Shim the engine mounts if necessary. Do not grind your rack and pinion for clearance. Anti-seize should also be used on the pinion spline. 

3. Connecting the power assist:

On power-assisted type G units, connect the hose from the servo port "L" to the left turn side of the rack cylinder (which is the head casting with the elbow fitting, not the driver’s left). Connect the "R" port on the servo to the right turn side of the cylinder, which is the plug end with the straight fitting. The "L" and "R" designations apply only if the servo is installed with its centering adjustment set screws toward the driver. If the servo is installed backward, left and right will be reversed. Upon initial startup this condition may cause the steering wheel to snap violently to full lock in one direction, so take care to get it plumbed correctly first. 

4. Adjusting tie rods:

Remember to install and adjust your tie rods so as to properly utilize the stroke of the rack. For example, on a typical dirt car (which requires more steering to the right than to the left) make sure the left tie rod is not too long or it will act as a travel stop against the rack housing, which will limit your steering to the right (in the case of power-assisted steering, you will be repeatedly driving the rod end against the housing with a thousand pounds force). To correct this situation, simply adjust the left sleeve shorter and the right sleeve longer by an equal amount. If you cannot adjust stroke interference out of the tie rods without causing an unacceptable bump steer pattern, then either (A) your rack length is incompatible with the distance between your control arm pivot points, or (B) your mounting bracket is too far to one side and should be cut off and moved. Incidentally, it’s not all that uncommon for one or both of these problems to exist on brand new race cars (See also step 6, below). 

5. Using bump-steer shims:

When adjusting your bump steer pattern by adding spacers under the tie rod ends, it is best to avoid spacing the inner ends any higher than absolutely necessary, as this merely increases the leverage acting to deflect them, especially in the fore-and-aft direction. Since this deflection only occurs under dynamic conditions (i.e., when actually racing) it will not be evident during bump steer measurement. It is thus possible to waste much time stacking spacers in pursuit of some theoretically ideal set of numbers and end up with a sufficiently elastic linkage as to render the numbers meaningless. 

Whenever possible the spacing should be done at the outer tie rod end. If the shimming amount gets beyond an inch or so, use steel spacers and tack-weld them to the steering arms. This will significantly reduce the bending load on the bolt. If you have to raise the inner tie rod end by more than 3/8", it’s best to shim under the rack housing itself. Remember, above all, that good geometry cannot be obtained without a stable and mechanically sound steering linkage ! 

6a. Making it turn left:

A frequently encountered problem with stock cars, especially on dirt, is the inability to initiate a left turn with the steering wheel. In order for the car to assume a fast cornering attitude, it first has to point to the left. If the car exhibits sluggish entry and will not rotate (especially on very short race tracks) it may have insufficient steering differential, or Ackerman. 

In order for any four-wheeled vehicle to roll around a corner, the inside front wheel must be steered at a greater angle than the outside wheel, because it has a sharper turn to make than the outside wheel. The tighter the turn, the greater this difference in steering angle. A simple test for Ackerman is to push the car around the parking lot (disengage one axle if you use a spool) while reaching through the window and turning the steer ing wheel to the left. If it seems less responsive than a street vehicle, or if the front tires chirp or skip, it probably has insufficient Ackerman. It may even have reverse Ackerman (see Fig. 1 below), which makes a car so unresponsive to the steering wheel that its effect on corner entry cannot be compensated for by any amount of bump steer, roll steer, or tire temperature, all of which happen too late. 

Most professional chassis builders incorporate steering differential into their front ends, one way or another. While with pavement cars it is often possible to ignore those details of geometry which involve steering to the right, in the construction of dirt late models sometimes both directions get ignored. In either case, the rack may wind up installed in the chassis without reference to the location of the steering arms (see Fig. 1). 

Frequently its installed position is subordinated to that of a radiator, harmonic damper, or other component, without the builder realizing that the steering geometry of the race car has been compromised. In the case of a used race car, somebody may have installed spindles with shorter steering arms, and, if the rack has not been relocated rearward by a corresponding amount, the Ackerman will be gone (assuming the car had any to begin with). 

A note: on fabricated spindles, Ackerman for left turns is often created by simply making the left steering arm shorter than the right, which gives a faster steering ratio to the left wheel. This makes the front wheels toe out when turned left, at a rate determined by the difference in length of the two steering arms. Unfortunately, it makes the wheels toe in when turned back to the right. 

On pavement this defect is usually irrelevant, since the toe-in won’t happen unless the car should rotate beyond the point of neutral steering and into the realm of counter steering into a slide angle. 

On dirt, however, counter steering is the prevailing condition. Considering the known effects of toe-out (stable state/push) and toe-in (unstable state/ loose), it’s easy to see that unequal steering arms tend to shift a cornering dirt car from one state to the other. If your car is twitchy in a slide, equal- ing out the steering arms may cure the problem. 

In stock car racing, a traditional fix applied to a twitchy car is to increase the static toe-out of the front wheels (crude, yes, but it almost always works). The following figures show what actually happens when this is done: 

Fig. 1 - The rack and tie rods are on a common centerline, and the steering arms point straight ahead. Everything looks all neatly squared up with the wheels straight ahead, but upon being steered three inches to the left (dotted lines), backward Ackerman is obvious! This car will not point left worth a damn, especially on dirt. These guys get to run the consi. 

Fig. 2 - In desperation, they adjust an inch of toe-out into the front end. Presto, the car gets in better, winning the consi and incidentally leading everybody to think that static toe-out is the answer. Is it? Not really! The static toe adjustment angled the steering arms outward by half an inch on each side, which created positive Ackerman, which was such a huge improvement in turning that it overcame the scrubbing effect of the toe-out down the chutes. 

Fig. 3 - The front end has now been improved by altering the spindles so that their steering arms are permanently diverged with the wheels still pointed straight ahead. This will generate positive steering toe (Ackerman) while eliminating the necessity for static toe-out. A further improvement has been made while they were at it: the rack has been moved an inch rearward, which not only helps generate Ackerman but, as a bonus, straightens the push-pull alignment of the tie rods as the steering angle increases. Compared with Figures 1 and 2, this is going to feel like a whole new ball game. Not only will corner entry be more positive, but it will now be easier to control the car in a slide. 

NOTE: It is entirely possible that a comparison of these diagrams with the front end layout of your car may suggest certain physical improvements. If so, they are worth doing. Getting rid of reverse Ackerman is one of the most dramatic and instantly rewarding changes you can make on a race car. 

6b. Improving slide-angle stability:

The foregoing examples dealt mainly with initiating a left turn on dirt (or, on pavement, maintaining a left turn). What about counter steering? The need for positive Ackerman still applies, since the function of the front wheels has not changed. They still have to steer the car through a turn regardless of the attitude of the car or the slip angle of the rear wheels.  In figure 3, both spindles were altered equally. Since Ackerman will now be generated identically to the right or left, the front end will be more stable under the transition from steering to counter steering.  Since counter steering involves steering to the right relative to the car, it’s easy to forget that the front wheels are still aimed around a left turn relative to the track and that the requirement for positive Ackerman geometry is still valid, if not more so. 

Fig. 3a - Although one car is steering left and the other right, their front tires are on practically identical paths (the wheel on the inside of the turn is always the one steered farther relative to the line followed by the car). 

Remember: the inside wheel must get around a shorter race track than the outside wheel, even with the car sideways. 

A further refinement possible on many cars is to straighten out the steering shaft routing. The steering shaft is frequently laid out as an afterthought (routed around structure, headers and so forth) and will often incorporate high u-joint angles. Excessive angularity causes nonlinear steering, which is a speeding up and slowing down of the steering ratio during rotation of the steering wheel. Depending on its relationship to the position of the driver’s hands, the cycle can be unnoticeable until a certain point is reached in the turning of the wheel, whereupon an unexplainable spinout occurs. If your car exhibits this tendency, it may be worth examining the steering shaft layout to see if there is room for improvement (there usually is). Any increase in mechanical smoothness and efficiency that it is possible for you to make in this area will pay dividends by broadening the driver’s useful range of steering input. The broader this range, the more forgiving.