Back one page          NEXT PAGE


Page Nomenclature

c.g.  = center of gravity

H = height of c.g. above datum, in.

µ = mu = coefficient of friction

SF = rollover safety factor, %

SUV = sport utility vehicle

T = wheel track, in.


Evaluation of a standard SUV

The next example shows that it is possible to build a high c.g. SUV which has sufficient safety factor, SF.

Case 3     
Standard Sport Utility Vehicle (rear drive)

Weight, curb 3600 lbs.
Load, max. 1150 lbs.
Wheel track, T = 68 in.
Empty c.g., height, H = 33 in.

This SUV is similar to Case 1, previous page, which scored unsafe. However, this SUV  has one notable difference: T is 10 inches wider. The extra weight of this SUV is due to more structure to accommodate its width. For the purpose of simplicity, we will assume that loaded vehicle H is 36 in., exactly the same as Case 2.

Calculating the safety factor, SF, it can be determined that is rated, "Marginal", with a SF of 27%. It should now be obvious to anyone, that compensating for a high c.g. is futile. "Marginal" may be good enough for someone who needs high ground clearance at the expense of passenger safety.

Evaluation of a high performance SUV

So just high wide must an SUV be to have the same safety factor as the lowly station wagon of Case 1? The final example shown below will be of interest to anyone planning on using an SUV for daily transportation of passengers.

Case 4     
High Performance Sport Utility Vehicle (all wheel drive)

Weight, curb 7200 lbs
Load, max. 3700 lbs.
Wheel track, T = 74.2 in.
Empty c.g. height, H = 33 in.

Using formula (1), the safety factor is 38%, the same as the station wagon of Case 1.

High c.g. is very expensive to overcome. Remember, the difference in H between Case 1 and Case 4 is only 8 inches! This SUV is larger than a Hummer®, aka Humvee! Auto designers know how to make a safe SUV; the only problem is a Humver® is huge, consumes much fuel and needs truck side lighting. Incidentally, the Rollover Safety Factor Chart produces the same results as the formula and shows Case 4 to be well within the light green zone.

Need for proving-ground testing

About the only cars that can approach the dark green (very safe) zone are low, wide sedans. It would be interesting to analyze the old American Motors Pacer®. It was wide and low, a seemingly clever design, but it did not sell well. Because the Pacer® was built on a very short chassis, it may present a problem with respect to longitudinal c.g. position at maximum gross weight. Studies would have to be done on it in order to ensure optimum balance. Also, since this car was a rear wheel drive, it would be essential that loaded c.g. was well forward of the longitudinal center of the wheelbase. I presume the Indianapolis-type racing cars would also do very well with c.g., but I would venture to say they considerably oversteer, thus making them a "bear" to drive.

Proof of stability must be established by means of a severe road test. Mathematical prediction is used first to design in as much stability as practical. It is beyond the scope of this presentation to cover all the factors of vehicle dynamics, however, it is generally accepted both from theory and practice that the most benign configuration is that of a front axle heavy car with front wheel drive and a low c.g. As far as I know, as a car approaches its dynamic limits, it becomes difficult to mathematically predict its exact road behavior because many variables become unstable; design-on-paper has its limits. That is the reason why the major manufacturers maintain proving grounds. I will suggest a way to test cars, and in particular, SUVs to verify their safety, at least in their tendency to tip over or go out of control. Remember, even a car with an optimum SF can still overturn if it slides off the road and "trips". So good emergency control as well as resistance to tipping go hand-in-hand.

Cornering-stability road testing


A 400 foot square flat concrete surface is needed. An airport tarmac will suit the purpose. At its center, a 100 foot non-elastic line is temporarily anchored so the other end can be held taut and used to establish a circular painted line. The resulting circle is 100 feet in radius and 628 feet in circumference. It is estimated that no vehicle tested will be capable of circling the circumference faster than one revolution per 12.6 seconds.

The test vehicle should be loaded to maximum gross weight, distributed normally. Tires should be inflated to factory recommendations and minimum fuel carried. The driver should be afforded all the protection available and emergency equipment should be standing by. The surface of the test area should be dry and swept. 

The trials will take place on the circle using several techniques.

  1. Both clockwise and counterclockwise rotations will be taken by the vehicle. Driving should commence on the circle by very slow acceleration up to the limit beyond which the car cannot be held centered on the line. The maximum speed can be measured by timing the circuit. Any untoward tendencies should be noted as well as subjective driver impressions.

  2. The driver can then repeat the same test, only applying wide-open-throttle (WOT) while circling at the maximum attainable speed.

  3. Then, the test is again repeated, only the driver applying maximum braking force at maximum circling speed.

  4. The second phase of testing is accomplished by approaching the circle at a tangent at a somewhat excessive speed, about 40 mph. The car is steered onto the circumference at minimum power and its behavior as to how it departs is noted.

I will now make some educated guesses as to the test results expected of several types of SUV vehicles. These opinions are expressed in Fig. 8.

(reference to above itemized list)

SF in red zone, nose heavy, rear wheel drive, conventional differential

Limited speed and moderate oversteering in (1),  must not perform tests (2), (3) and (4) as car will overturn.

SF in red zone, nose heavy, rear wheel drive, limited slip differential

Do not attempt trials. Fails all tests. Will overturn. 

SF in orange zone, nose heavy, rear wheel drive, conventional differential Moderate oversteering in (1),  must not perform tests (2),(3) and (4) as car may overturn
SF in orange zone, nose heavy, rear wheel drive, conventional differential Do not attempt trials. Overturning a possibility.

SF in yellow zone, all types

Cannot predict outcome. 

SF in green zone(s), nose heavy or balanced, rear wheel drive, conventional differential Moderate oversteering in (1), severe oversteering and possible loss of control in (2), spinout in (3), understeering  then departure from circle in (4). Remains upright.
SF in green zone(s), nose heavy or balanced, rear wheel drive, limited slip differential Oversteering in (1), loss of control in (2), spinout in (3), understeering and departure from circle in (4). Remains upright.

Fig. 8

I remind the reader that anytime an oversteering situation is incurred, there exists a real possibility that total loss of control will follow. This is because oversteering is very difficult to handle; racing car drivers have the skill to do it, but the average driver usually does not. Instead, it is usually during an accident avoidance maneuver or accidentally entering a decreasing radius turn too fast that he discovers his vehicle has this dangerous property and unfortunately leads to loss of control. Even if the vehicle does not upset, a very hazardous situation is set up. It is far better for any passenger car sold to the public to have the forgiving nature of understeering. In that case, usually the car will  simply drift out to a larger radius of turn, while maintaining directional control. Recovery usually follows from what would almost certainly be an adverse event with an understeering car. Most front wheel drive passenger cars are understeering, but relatively few SUVs are front wheel drive. 


Back one page         Top of page          NEXT PAGE