1. Home
2. Published pieces
3. 911 suspension

Porsche Suspension Testing

Dialing-In The 911

By Greg Raven
Photos by C. Van Tune
VW&Porsche magazine
1987

The 911 has always been recognised as a super sports car. It is quick, fast, has great brakes, and handles well. In addition to all the performance built into the car at the factory, there is a seemingly limitless amount of extra performance that can be extracted by those willing to take the time to do so. Witness the 934, 935, 959, and 961 as examples of just how far the 911 theme can be pushed.

Back in April 1986, we teamed up with Automotion to do some suspension testing and improvements on the 928. It worked so well that we decided to try it again. Only this time, our subject was the 911.

We knew before we even started that the 911 was going to be capable of much better handling that it comes with out of the box. First off, the ride height is set at the factory not for optimum handling but to meet the headlight height requirements of the United States.

Second, given the predominance of relatively unskilled drivers in the United States and our national predilection to litigation, every car sold in this country is set up to understeer when the driver enters a corner too fast. The feeling is that if you understeer your way into an accident, it is your fault; if you oversteer your way into an accident, it is the automaker’s fault. In the 911, Porsche has a peculiar situation because with the rear engine design of the 911 snap oversteer can be a big problem. Our challenge would be to balance this oversteering tendency against the need for the more neutral handling demanded in an autocross car. With skilled drivers at the helm, we were confident that this could be done.

For our test course, we again borrowed the same large parking lot at the Alameda County Fairgrounds that we used for the 928 testing, setting up the same course. The Golden Gate Region of the PCA uses this course layout on this particular piece of pavement for their autocrosses, so we had an idea of what to expect from a well-set up car.

From the start/finish line, the course immediately makes a 90 degree left turn into a short straight. At the end of the straight there is a 45 degree left turn that is the entrance to a chicane. The exit of the chicane is the entrance to a 180 degree right-hand sweeper that empties out onto another short straight. At the end of this straight the course dives into a 90 degree right-hand turn, followed almost immediately by two 90 degree left turns. This brings the course back to the start/finish line.

The car used for the testing was a 1983 911SC with 46,000 miles on it; owned by Steve Rinkle, sales manager at Automotion. The car had been prepped by Denny Kahler, owner of Kahler’s Import Shop in Dublin, California. Course layout, timing, and the myriad of other details that accompany a project like this were attended to by John Peichoto, an official of the Porsche Club, and Tom and Marjorie Green, the owners of Automotion. Driving duties were borne by Dwight Mitchell (suspension consultant and owner of Autosport Technology in Sacramento, California), and C. Van Tune.

Phase I

When Steve bought the car it had stock suspension, tires, and wheels, and the previous owner had lowered the car to California Cruising height. Although we did not measure the ride height, it was obvious the car was much too low.

Because of the large number of cars that you see driven daily on the street in this condition, we decided to test it as-is, to establish a (admittedly slow) baseline time. Inflating the Pirelli P7 205/55 and 225/50 tires to 40 psi all around (cold), we set out for the autocross course. We shaved the tires for these (and all subsequent) tests to prevent the tires from chunking.

Our best time was 58.739 seconds. The most noticeable feature was the exaggerated oversteer. This was caused partly from the car being too low in the rear and partly from the rear suspension bottoming out.

Phase II

The first order of business was to get the car off the ground so the suspension could begin to work properly. Automotion’s philosophy is to set the car as low as possible (so the center-of-gravity is low) and then optimize the suspension to that height. To this end, Denny raised the rear of the car ¾ inch, and realigned both front and rear suspensions. In the rear, the settings were 2 degrees, 30 minutes camber, with 10 minutes of toe-in at each side. In the front, we were shooting for about 1½ degrees negative camber but we could only get 50 minutes. This was an indication that we were right at the lower limit in ride height. Front toe-in was set at zero, and caster was set at 5½ degrees.

At this time we also changed to Koni adjustable sport gas shocks set ⅔ hard (2 clicks off full soft). Because of the lower ride height, the shock absorber bump stops were cut in half and the harder half reinstalled. Because we were not sure that this was going to be our final set-up, we did not corner balance the car at this time.

Test One of this phase was run on the original Pirelli tires. Although the tires were fairly well worn out, we were able to tell that the car was more stable and easier to drive at speed. There was still a great deal of understeer evident, however, in the 58.198 second time.

For Test Two we installed Yokohama A008G tires on BBS forged wheels. The A008G is a revised design of the A008 with stiffer sidewalls, wider (and different) profile, and improved belts for better transitions. The BBS wheels were much wider than the stockers, running at 7.5 inches in the front and 8.5 inches in the rear. Not surprisingly, the time dropped significantly to 57.306 seconds.

For Test Three we started working on the understeer by using two pounds more pressure in the front tires than in the rear. The car still pushed, but it was better, as the 57.288 time will attest.

Phase III

Satisfied that the basics were right, the final step was to optimize the suspension for the ride height. Weltmeister anti-roll bars, 22mm front and rear, were installed to complement the Sway-A-Way 22mm front and 28mm rear torsion bars. To make minor adjustments more easy, Sway-A-Way adjustable rear spring plates were also installed at this time. To eliminate bump steer in the front, Weltmeister steering rack spacers were installed, as were Automotion front turbo tie rods for more precise handling.

An important part of any 911 suspension change is corner balancing the vehicle. Because we were fairly satisfied with the results so far, we decided to corner balance the car at this step. The final weights were: LF — 580, RF — 565, LR — 1000, and RR — 980. This resulted in a ride height of 160mm in the front and negative 5mm in the rear. Stock U.S. ride heights are 99mm in the front, 37mm in the rear, while stock European ride heights are 108mm in the front, 16mm in the rear.

After disassembling the suspension the alignment had to be redone. In the rear, the settings were 2 degrees, 10 minutes camber, with 0 toe-in. In the front, we were able to get closer to our target setting with 1 degree, 35 minutes of camber. Front toe-in was set at zero, and caster was set at 6 degrees, 30 minutes.

Test One in this phase showed that the chassis was very well balanced, and it seemed to be one of the best 911s we had ever driven. The handling was very stable and predictable in a variety of corners. There was a feeling of confidence in driving the car, which was quite a contrast to the apprehension that many feel when auto-crossing a stock 911 suspension. With the Yokohama A008G tires mounted on the BBS forged alloys, the best time was 55.676 seconds.

For Test Two it was time to try out the Yokohama AVS dry tires. With no other change the car immediately felt improved, with more control, better response in the chicanes, and better power to the ground coming out of the sweeper. The best time was 54.992 seconds.

Because the AVS tires seemed to be sticking pretty well, we decided to try to tighten up the front anti-roll bar to improve the transition through the chicane. It worked. The time dropped to 54.506 seconds.

Conclusion

Through careful testing we had radically transformed the handling of Steve’s 911SC, knocking over 4.5 seconds off the autocross time without destroying the car’s utility as a street machine. If the goal had been to put together a car to be used solely for autocross, we could have continued on further from here by increasing rear roll stiffness, transferring some weight to the front, etc.

As it was, everyone was satisfied with the results. Even the sky waited until we had finished the last lap before letting go with the rain it had been threatening all day.

Visit Automotion on the web.