Minimizing chassis roll in corners improves performance on any car used on the track, but it is especially important on cars equipped with strut suspensions.

The suspension setup for a strut-equipped car is very much like that of a double wishbone “A-arm” suspension. You adjust camber and tire pressures using tire temperatures. Then, by using average tire temperatures, you can balance the handling and maximize total vehicle traction in the same way. But there is a trio of major differences that can affect total vehicle traction and drivability.

The first is simple. Strut-equipped cars typically have more severe camber-change curves during suspension travel. This means that the angle of the tire tread compared with the ground at the tire contact patch is causing a portion of the contact patch to lose contact with the track surface, reducing traction. This is a transient problem, meaning that the more the chassis rolls in a corner, or compresses or rebounds over bumps, the more or less the tire contact patch is gripping the track surface. The older the car, the more likely this will be true.

So total grip of the car is changing as the car is turned into a corner, over bumps in the corners and during acceleration exiting a corner. Less traction means slower cornering speeds and less corner-exit acceleration. It also makes the car difficult to drive. Since traction or grip levels can be changing constantly in a corner, the handling balance also can be changing, causing a car to change from understeer to oversteer very abruptly. It is also difficult to feel what the car is doing. In other words, the car may turn in with good grip, but as chassis roll increases, grip is reduced due to camber change and the car may understeer if the problem is at the front, oversteer if the problem is the rear — less likely — or start sliding off the track if both ends lose grip equally, which is rare.

There are two solutions to the camber-change problem. First you can change the suspension geometry, which is the best solution but difficult and/or expensive. Second, you can limit chassis roll. Camber change is directly proportional to chassis roll, so reducing roll reduces camber change. Stiffer springs are not the best choice for this, but stiffer antiroll bars are.

Limiting chassis roll to no more than 2 degrees reduces the camber-change problem to a minimum. In fact, this is one situation where a stiffer front antiroll bar can reduce understeer. This was a trick used in stock-class autocrossing where rules allowed only the addition of a front antiroll bar. By reducing chassis roll, more of the tire contact patch was keep flat on the track surface, allowing more contact and more grip. Understeer was reduced, even though adding roll stiffness to the front of a car increases front roll couple distribution and should cause more understeer. The key, as always, is keeping as much of the tire tread in contact with the track surface.

The front tires on front-drive cars steer and provide traction for 60 to 70 percent of the cornering and braking, and all of the acceleration. Minimizing chassis roll is key to getting all of the tire contact patch working on the racing surface.
The front tires on front-drive cars steer and provide traction for 60 to 70 percent of the cornering and braking, and all of the acceleration. Minimizing chassis roll is key to getting all of the tire contact patch working on the racing surface.
This Honda in midturn has about 2 degrees of chassis roll. Note that the outside front tire, which is the most heavily load tire in a turn, still has a little negative camber. This is the optimum situation for maximum cornering speed.
This Honda in midturn has about 2 degrees of chassis roll. Note that the outside front tire, which is the most heavily load tire in a turn, still has a little negative camber. This is the optimum situation for maximum cornering speed.

The second problem relates to the length of the strut when a car is lowered. If a car is lowered with stock struts at stock length, there is too little bump travel for good handling. The shock will bottom out, causing serious handling problems. For a road-racing or track-day car that is modified for the track, the struts must be shortened or replaced if the ride height is lowered. Today, this is fairly easy with many aftermarket suspension companies offering modified strut and spring combinations and, even better, coilover struts with adjustable spring perches and even adjustable camber plates.

Properly designed strut suspension systems using coilover springs and adjustable ride heights like these from Progress Technology make setup on a strut car much easier.
Properly designed strut suspension systems using coilover springs and adjustable ride heights like these from Progress Technology make setup on a strut car much easier.

The first strut car I encountered was 25 years ago. It was a Datsun 240Z running in the Improved Touring S class. The car owner was struggling with setup. He asked me to drive the car since I had been his racing instructor. He was concerned that it was his driving. After only two laps around the 2.5-mile Willow Springs International Raceway “big” track, I came into the pits. I asked the owner if he wanted the good news or bad news first. He said the good news first. I told him he was a much better driver than he thought, just to keep the car on the track. The bad news was obviously that the car was horrible.

At first I thought the problem was extreme bump steer, which was part of the issue. But after more checking and contemplation, I realized that the Z was bouncing off the bump stops, causing the spring rates to go sky high, first at the front, then at the rear, causing the car to dart over bumps in the turns and to transition from big understeer to big oversteer and back again. Just keeping the car on the track was a challenge. The struts were stock length, but the car was lowered a couple of inches. Additionally, the owner was running on a 225 cross section DOT race tire instead of a 205 section tire. While the extra tire width should have increased traction, the camber change (more positive camber) was causing only the outside section of the tires to be working on the track surface. Narrower tires reduce the effect. But the big issue was the bottoming.

To cure the problem, we shortened the strut tubes, added adjustable camber/caster plates and rebuilt the Koni shocks because they had bottomed so badly the valving was blown out. We added stiffer antiroll bars and used shims on the outer tie rod ends to align the tie rod end with the ball joint to eliminate bump steer. The net effect was a transformed car that was now easy to drive and very competitive in the ITS class.

The third and final issue with struts is also the most difficult to resolve. It also relates to suspension geometry. The problem is the potential extreme lateral movement of the roll center, especially at the front on front-drive cars. Newer front drive cars are less prone to this issue, but older cars are seriously affected by this. Some history may help explain the issue.

In the early days of modern suspension design, vehicle dynamics was not well understood, partly because those who understood dynamics rarely shared their knowledge. In the 1970s, General Motors published a book, which included information on suspension design and dynamics. This was the basis for several books on racecar suspensions in the years following its publication. Within those books, an assumption was made that the roll center of a suspension system was fixed along a vertical axis. This meant that the roll center could not move laterally but only vertically. The correct element of this thinking was that the vertical movement of the roll center on an independent suspension system, which is called the “jacking effect” altered the loading on the tires. In other words, a small amount of tire traction was being used to jack the chassis up (or down in some cases). So designers tried to create suspensions that minimized the jacking effect as well as keeping the roll center very low, within a inch or two from the ground. At the time this was mostly front suspensions on rear drive cars, usually with a solid rear axle.

By the mid to late 1980s, Wm. C. Mitchell, with help from Carroll Smith, created the first — to my knowledge — suspension geometry analysis software program call “Racing by the Numbers.” This was a DOS-based program for early PCs. But it was brilliant. Mitchell and Smith understood the parameters of suspension geometry, and those elements were built into the software. I was fortunate enough to use the first version of the software while analyzing and designing suspensions. I consulted and tested for several aftermarket suspension companies, mostly conducting skid pad and slalom testing.

The Performance Trends Suspension Analysis software illustrates the extreme movement of the roll center on a strut car. This screen shot shows a double “A” arm system, but the software also works with strut suspensions.
The Performance Trends Suspension Analysis software illustrates the extreme movement of the roll center on a strut car. This screen shot shows a double “A” arm system, but the software also works with strut suspensions.

Around 1990, a racing friend contacted me to analyze the suspension geometry on a car he had acquired. He had a long history in the IMSA Firestone Firehawk Series running Camaros. But he made the move to the IMSA International Sedan Series when he purchased the ex-factory Shelby CSX. The CSX was based on the front-drive Dodge Shadow coupe, which used a turbocharged 2.2-liter four-cylinder engine. This was my first experience with a front drive racecar and with a turbo. The car was not really competitive when run by the Shelby factory team, so our first goal was to track- and skid-pad-test the car.

My first quick lap wasn’t. The turbo boost came on so abruptly that the car literally “jumped” off the track at the exit of Turn 1 on the Streets of Willow track in Southern California. The abrupt torque steer was quite a surprise. But beyond that, the car was very difficult to drive since it lacked any reasonable steering feel. Plus the car reached only 1.01 g’s on the skid pad on BF Goodrich DOT race tires, the best available at the time. So it was time to analyze the suspension geometry with Mitchell’s “Racing by the Numbers.”

After plugging in all of the dimensions and data into the computer, we calculated correct spring rates, antiroll bar rates and analyzed the geometry. By limiting chassis roll with stiffer spring rates and much stiffer bars, we knew that we would reduce some of the car’s issues. But when we looked at what was happening with roll center movement, even when limiting chassis roll to less than 2 degrees, we were somewhat mystified by what we saw. Vertical movement was less than 2 inches at two degrees chassis roll, not great but manageable. But lateral movement was shocking. The roll center was moving about 2 feet outside the track width of the car — or more than 5 feet from the chassis centerline. We rechecked the measurements, ran the numbers again and had the same result. We could not find any information on how this might affect handling, but knew it must not be desirable.

When the chassis rolls, it pivots around the roll center along the roll axis, which is an imaginary line running through the front and rear roll center. This car, like many front drivers, had a solid rear axle with lateral locating device, so the rear roll center stayed on the chassis centerline. But the front roll center would begin to move to the outside of the car as soon as the chassis began to roll, meaning that the roll axis was swinging wildly from a pivot point at the rear axle centerline to well outside the chassis at the front with 2 degrees of chassis roll. To really understand this, you need to understand another important part of the puzzle.

One of the factors affecting the amount of chassis roll is the length of the moment arm from the roll axis to the car’s center of gravity. When a car has high roll centers, the roll axis is closer to the center of gravity, which is always less desirable than low roll centers due to camber change curves and other variables. But with low roll centers, the amount of chassis roll increases since the moment arm, or amount of leverage from the roll axis to the center of gravity, increases. This is why much stiffer front and rear antiroll bars are important to reduce chassis roll at high cornering loads.

The goal is to minimize camber change, which causes the tire contact patch to partially lose contact with the track surface, reducing cornering speeds. Camber change is usually more abrupt and more severe on a strut suspension, especially at the front end.
The goal is to minimize camber change, which causes the tire contact patch to partially lose contact with the track surface, reducing cornering speeds. Camber change is usually more abrupt and more severe on a strut suspension, especially at the front end.

Using stiffer springs will work, also, but then the spring’s primary job, to keep the tire contact patches planted to the racing surface, is compromised, especially on bumpy race tracks. When the roll axis stays very close to the vehicle centerline as the chassis rolls, the weight transfer from inside to outside in corner will put the outside suspension into bump and the inside suspension into rebound travel. But when the roll axis swings laterally because of roll center migration laterally, everything changes.

When the roll center moves laterally, the roll axis swings. This makes the length of the moment arm from the roll axis to the center of gravity longer, causing the leverage on the suspension to become greater. If the roll center moves to the outside of the car (away from the turn direction), the roll of the chassis causes more compression of the outside suspension and less rebound, or extension on the inside. As the roll center moves to the outside, near the outside tire contact patch, this trend increases until the roll center is outside the track width of the car. Then the weight transfer to the outside tire is trying to lift the inside tire off the racing surface.

When the roll center moves to the inside, the weight transfer to the outside is trying to lift the inside tire even sooner. In both cases the outside tire is overloaded and energy from the tire traction is being used to compress the suspension on the outside and lift the inside tire. Both mean a loss of cornering speed.

What happens during the transition — corner entry and exit — is even worse. As the roll centers move and the roll axis swings, the effect of the weight transfer is constantly changing. This changes the vertical load on the tires while the chassis rolls. The changing load is not linear, as it would be with a stable roll axis. So the driver has a very difficult time “feeling” what the tires are doing. Combined with the loss of traction, the lack of feel makes the car slow and the driver has a hard time judging handling balance and grip levels.

On the Shelby CSX, we played with lower control arm pickup points until we were able to keep the lateral roll center movement to less than 2 inches at 2 degrees of chassis roll. Combined with the other changes listed earlier, the car was transformed. After tuning the camber and adjusting the rear antiroll bar rate, the car topped 1.09 “G”s lateral acceleration on the skid pad and was more than 2 seconds faster around the Willow Springs street course. Additionally, the steering feel improved, making the car much easier to drive at the limits of tire adhesion.

The roll-center movement and associated camber-change curves are a serious issue on most strut-equipped cars. This is more true for older strut-equipped cars. Explaining how the roll center moves, and its effect on traction and driver feel is difficult. Illustrating what happens is even more complex. To really find out what is happening on a strut car requires a computer program like Mitchell’s “Racing By The Numbers” or others listed in the sidebar. To change pickup points is a difficult and costly proposition. But that is not necessary in most cases.

Many strut-equipped cars can have major handling gremlins, but the solutions for improved performance are not much different from any production-based car moving from the street to the race track. The most important element is to tune the chassis and suspension based on what the tires need to generate the maximum amount of traction. In the case of a strut car, reducing chassis roll is even more important than with a double wishbone suspension-equipped car.

The roll axis is an imaginary line passing through the front and rear roll centers. The distance from the roll axis to the CofG affect the amount of chassis roll.

Since most front-drive cars use struts, it is important to remember that the front tires are doing 75 to 80 percent of the work, so getting the most traction from the front tires is an important task. If you run a strut car and have a hard time feeling what the car does at the limits of adhesion, or the car just seems to lack grip, then hopefully, you will have found some answers here.

The geometry of the control arms determines the location of the roll center. Roll centers will move vertically and horizontally during suspension travel.
The geometry of the control arms determines the location of the roll center. Roll centers will move vertically and horizontally during suspension travel.

“Cliff’s Notes” on Strut-Tuning

Reading an entire feature on tuning strut suspensions is a lot to take in. For review, here is a bulleted list of the steps necessary to minimize the possible negative traits of a strut car.

  • First, 90 percent of the setup on a strut car is the same as any car. Use tire temperatures for optimizing setup.
  • If you lower the car, and most likely you will, be sure to use shorter strut inserts (shocks) and shortened strut housings to avoid bottoming. Aftermarket kits from quality suspension manufacturers make this easy.
  • Chassis roll while cornering is the major culprit, so reducing chassis roll is important. This is also important to control camber change and improve traction, especially during the transitions of corner entry and exit. Chassis roll is reduced by increasing roll resistance (stiffer spring rates and antiroll bar rates).
  • Most of the roll rate increase for the front suspension should come from the antiroll bar. Some of the roll rate increase can be provided by the springs. But if the front springs are too stiff, especially on a front-drive car, the front suspension frequency (a measure of the spring rate combined with sprung weight on a wheel acting at the tire contact patch) will be too high. A frequency that is too high will cause wheel spin and understeer over bumps.
  • The front antiroll bar needs to be stiff enough to limit chassis roll to less than 2 degrees. Even less will be better. Exceeding 2 degrees of chassis roll allows too much camber change and possible lateral roll-center movement.
  • If possible, use a front antiroll bar with some adjustment for rate. This will allow some tuning of front-to-rear roll resistance (roll couple distribution) so that a near-neutral handling balance can be achieved.
  • A rear antiroll bar is a necessity for tuning handling, and it must be adjustable (or have different bar rates available) so that roll couple can be adjusted for near neutral handling balance.
  • For proper tuning of camber and caster, it is necessary to use an adjustable upper strut mounts (spring perches) that allow camber and caster adjustments.
  • Several aftermarket suspension companies make strut kits for racing and track car applications. In most cases, these kits are engineered to lower the ride height with the correct length strut housings and inserts, the correct spring rates and antiroll bars that are either adjustable or engineered to provide a near-neutral handling balance.

We found this video on YouTube. It’s a simple animation of what happens to a vehicle’s camber throughout the range of a strut’s motion.

Resources

The links below go to companies that provide suspension geometry software. Racing Aspirations is a free online calculator.

http://www.racingaspirations.com/suspension-geometry-calculator

http://performancetrends.com/SuspAnzr.htm

Suspension Glossary

Here are a few terms you’ll need to be familiar with to understand the finer points of tuning a car with strut suspension. Or any car, for that matter.

Bump steer  — Toe change (in or out) caused by a disparity between the geometry of the front suspension compared to the steering geometry. If the arc of the steering tie-rod does not match the arc of the front suspension, one or both front tires are forced to steer during suspension travel. This is most apparent over bumps and dips, especially over one-wheel bumps. Bump steer can cause the car to dart over bumps and also causes tire drag, which reduces cornering speed and accelerates tire wear.

Center of gravity  (CofG) — The point where the mass of the vehicle is considered to be centered. The lateral point is where the car is balanced left to right. The longitudinal point is where the car is balanced front to rear. If a car has 50/50 weight distribution front to rear and left to right, the center of gravity is in the midpoint of the wheelbase and the midpoint of the track width. The center of gravity height above ground is more difficult to measure, but is usually very near the centerline of the engine crankshaft above the ground line. The higher the CofG is above ground the more weight transfer there is.

Camber — The tilt of a tire when viewed from the front.  Camber is positive if the top of the tire is tilted to the outside and negative if the tilt is to the inside.  Since most suspension systems gain positive camber during bump (compression) travel, and the outside tire goes into bump during cornering, some amount of negative camber is needed to offset the camber gain and keep the tire contact patch flat on the road surface during cornering.

Moment arm  — Leverage applied to a point, like a torque wrench twisting a bolt or nut. The nut is the pivot point and the moment arm is the length of the torque wrench. A 50-pound force applied to the end of the one foot arm creates 50 pound-feet of torque or rotation. A 50-pound force applied to the end of the two foot arm creates 100 pound-feet of torque or rotation. When the center of gravity is above the roll axis, the forces created by cornering (from the tires) act at the center of gravity. The moment arm is from the center of gravity to the roll axis. The longer the arm, or distance between the CofG and roll axis, the more chassis roll will occur. If the roll axis passes through the CofG, then there is no chassis roll. But this would take a very undesirable suspension geometry.

Oversteer/understeer  — Understeer occurs when the front tires lose traction before the rear, also called “pushing.” Oversteer occurs when the rear tires lose traction before the front, also known as “loose.” The NASCAR definition is: Understeer occurs when the front of the car hits the wall; Oversteer occurs when the rear of the car hits the wall.

Roll axis  — An imaginary line passing through the front and rear roll centers. The distance from the roll axis to the CofG affect the amount of chassis roll. It has no effect on the amount of weight transfer while cornering, but it can affect where the weight is transferred, front vs. rear.

Roll center  — A point about which a suspension rotates. The geometry of the control arms determines the location of the roll center. Roll centers will move vertically and horizontally during suspension travel.

Roll couple distribution — The total amount of roll resistance present in a car. Where roll couple is the total amount of resistance to body roll provided by the springs and antiroll bars at front and rear, roll couple distribution is the amount of roll resistance at the front relative to the amount at the rear. Changing the roll couple distribution balance of the car changes its handling balance.

Roll rate — Increasing roll rate reduces chassis roll and therefore reduces camber change. The antiroll bar does this best because it does not affect traction over bumps as much.

Spring rate — The load, measured in pounds, it takes to compress the spring one inch. The factors affecting the rate of a spring are the diameter of the spring, the number of active coils and the diameter of the spring wire. Larger diameter springs reduce the rate, few coils increase the rate and thicker diameter wire increase the rate.

Comments
Images courtesy of Progress Technology Photo, Performance Trends Suspension Analysis Image and Don Alexander