By Kiyoshi

Those that are familiar with the Formula One series of 1992 know how dominate a single team with a lead driver was. I am refer to the Williams Renault car driven by Nigel Mansell. With four races remaining in the season Mansell clinched the driver's championship, shattered records, as the Williams team finish 1-2 for an unprecedent number of victories. Why? Was it the superiority of Mansell and Patrese? One suspects the truth was stated by Mansell prior to the Japanese Grand Prix where he stated to the press with reference to Alain Prost (who replaces Mansell at Williams) that, "Even a puppet could win in a Williams". Clearly it was apparent all season that the Williams Renault was a far superior package than even the once mighty McLaren Hondas driven by Ayrton Senna. Mansell was simply stating the obvious, and what every team on the GP circuit had known since the opener in Brazil.

What was it about the Williams that made so invincible? Aerodynamics? Re-active Suspension? The Renault engine? Traction control? Semi-Automatic gearbox? It is easily argued all of the above factored into a total dominating package.

Something else happened this F1 season. The lowly (just look at the state of the team at the end of 1991) Lotus Team came from cars that were in jeaprody of not qualifying to cars that were challenging the more powerful Benettons and McLarens at the last few races.

It is no surprise then to find both Lotus and Williams using some form of hydraulically controlled computer monitored suspension control system - Active, Semi-active or Re-active suspension.

HISTORY
In 1981 FISA (all knowing and god-like) introduced a rule which forced cars to run 6 cm above the ground. The rule was created in an effort to drasticallly reduce the downforce created by the tunnels of the side pods (ground-effects). The rule was enforced by stewards in the pit lane measuring ground clearance. It was impossible to accurately monitor the cars while they were at speed on the track. Racers being the worlds best and most efficient loop hole locators quickly found a huge one. Gordon Murray of Brabham quickly put a smal hydro-pneumatic jacking device in the Brabhams so that when the car entered the pit lane the driver flipped a switch and instantly the car was raised to the mandatory 6cm clearance. Once out of the pit lane the driver flipped the switch in the opposite direction and the car dropped to the ultra low stance that allowed the side skirts to seal out air to the ground effects tunnels.

Chapman at Lotus followed suit but, saw that much more could be gained if ride height and the attack angle of the tunnels could be kept at their optimum. Thus Chapman outlined his thoughts concerning active suspension. By 1983 the first serious work on active suspension emerged from the Lotus garage, the Lotus 92. The components were commonly called hydraulic suspension and electronically controlled anti-roll bars. Peter Wright of Lotus Engineering was the chief behind these developments. Wright recalls, "We started with active suspension in the early 1980s, and we first raced the system in 1983 at the Brazilian and Long Beach Grand Prixs. That was a fully active system. It was fairly crude and heavy. It was really and experiment - to see what was involved in running active."

It is interesting to note that this first system was first driven in competition at the '83 Long Beach race by the first driver to clinch a world championship in a semi-active suspended car, that being Nigel Mansell. At the time when Lotus decided to set aside the system for further development, Mansell was quoted, "I still believe in the long-term possiblities, but it needs more development."

WHAT IS IT?
According to Lotus' Wright, "The system is able to change the attitude and ride height of the car according to certain control laws." The benefit is that at any given moment an active suspension system can optimize the suspension. To do this an active system can control three major areas: 1) The system can control ride height according to changes in weight and in aerodynamic loading. The system can react to internal loading, such as roll and be made to create roll or no roll. 2) The system can stop the car from pitching under braking or acceleration. 3) You can actually control the springing/damping and the hub motions of the car.

So called Re-active systems can perform only the first function. A Semi-active system performs the first and second functions, but only a fully active system actually achieves all three. Controlling the springing, damping and hub motion is obviously requires the most sophistication and is the most difficult to achieve. If it does not work there is no suspension at all! Wright, " With a semi-active system, if the computer says: 'I will do nothing at all,' you still have suspension." Wright continues by saying, "Those are great differences. There ar others, but these are really in the control laws - what you are asking the system to do. That is an area where it is impossible to know what other people are doing. You can only look at videos and guess."

What Wright is saying is that with a semi-active system there is a conventional passive suspension, springs and dampers. Whereas a fully active suspension, the system is capable of sensing bumps and nearly instanteously reacts. It does this so quickly that it's almost as if the suspension anticipates the bump. Compare this to a conventional spring/shock system and you'll find the conventional system extremely retarded, reacting with seeming casualness.

MORE HISTORY
Between 1983 and 1987, the Lotus active suspension disappeared (from F1). According to Wright, "The system was being developed for road car use by Lotus Engineering." Many GGLC members may recall seeing videos of an active suspended Excel actually 'banking' into corners and running a slalom virtually flat.

In December 1986 Williams tested an active system developed by Automotie Products. The car was tested by Mansell, Piquet and Jean-Louis Schlesser.

In 1987 Lotus had again embarked upon using an active system in F1. Ayrton Senna scored and victory at Monaco, the first for a fully active car. Senna backed his victory with another at Detroit. At the Italian GP Williams introduced a semi-active car driven by Piquet, developed by Frank Dernie, Piquet won.

Again Wright, "The Lotus was again fully active and relied on computers to simulate the spring/dampers units. The Williams was semi-active - or re-active - as it is now."

However Lotus returned to 1988 without active suspension. "In 1987, Lotus Engineering put the system on the car and funded it," explains Wright. "It was the decision of Lotus Engineering not to continue funding it in 1988. Although we were prepared to supply it to Team Lotus, we were using a lot of key people who were also needed to support quite a big active suspension business at Lotus Engineering, working for customers. Therefore, we could not justify the expenditure as well as the resources. If Team Lotus had wished to go on with it, and had been prepared to fund it, we would have continued. It was a Team Lotus decision."

By halfway into the 1988 season other teams were experimenting with semi-active. Both Ferrari and Benetton tested systems. But, then Williams pulled the plug on their system. Having lost Honda turbo engines and switching to much lowered powered atmospheric Judd V8s, the cars did not have the extra power needed to be competitive and drive the system.

Which brings us to the beginning of the 1992 season where Williams brought their updated re-active system and by SPA Belgium Lotus was running and semi-active system. "The system on the 107 is not a fully active system," says Wright. "We are using a conventional spring/damper unit, whereas Williams is using a hydro=pneumatic spring with an integral damping system. Otherwise, I believe the two systems are functionally similar although, obviouly the control laws are probably different in may ways." It should be noted the 'control laws' to which Wright refers are the algorithms or programing for the computer controlling the system.

Wright continues, "The Lotus system is quite simple and small. We really have gone for the simplest and most economical system we can devise in terms of weight and power consumption. It is as simple as we can make it, so that it is quick to develop and to achieve good reliability with. It uses many of the components and the knowledge that was learned on fully active systems, but simplified to be semi-active."

"The system we are using now is no more complex than a fuel injection system. It uses high-pressure feeds, control valves, sensors and a computer. The computer is reasonably sophisticated, but not sgnificantly more complex than any of the other computers in the car. The hydraulic/mechanical side is 1950s technology in concept."

"The sophistication is in the control laws. We perhaps think it is simple because we ahve been making control laws since the early 80's. The ones we use on this system are a subset of a fully active system, so we are coming back from a much more complicated set of control laws. In terms of reliability, it is a set of mechanical problems like any other. If you can make a fuel injection system that functions correctly and doesn't leak, there is no reason why you can not make an efficient hydraulic system."

"I do not subscribe to teh high-cost, high-complexity theory fo active suspension. The only real cost is in testing and developing. the actual components are not expensive by F1 standards - in relation to engines, for example."

"We have a fully active system still under development, and we plan to fit it on the 107," Wright admits. "We wanted to get the system on the car from the start, but we didn't want to hold up the development of the car, so we opted to so semi-active."

"It must be faster, because it is more complicated (though not necessarily heavier), although it is more difficult to run. We want to get to know, to calibrate and understand the 107 so that we know what we want to do with it. The active is only a system that has to be matched and coordinated with the car to produce the right performance. we have to understand the car before we go to the next stage. I think we will get quite a lot of the performance potential of active from the system we have on the car now, but I believe there is more performance to come from the fully active."

To maximize the benefit of an active system the ground clearance is critical. Consider that the one the major benefits of an active system is the reduction on reaction time of the suspension. This is a critical factor for the modern race car whose aerodynamic efficiency is tied to ground clearance and attack angle. Thus measuring ground clearance is extremely important. On this subject Wright says, "There are a number of techniques, but I will not go into detail about how we (Lotus) do it. There are some very good little laser ride height sensors that are very sensitive, and that we use for research purposes, but we don't have them fitted to the racing cars."

"People have talked about acoustic systems, as based on the Polaroid range finder. We have done some testing with acoustic systems, but we are not very happy with the results. I think the best ride height sensor at the moment is the laser."

THE COMPUTER
Obvious the processor speed is important. It is the brain that must receive the inputs from the sensors, run calculations, and then output signals to the hydraulic actuators and valves in a short enough increment of time that the reaction is meaningful. Thus the speed necessary for a fully active suspension system computer depends on what you need to do, but Wright is certain that everything required by such a system is feasible.

"If you just want to control fuel level, it would require very low frequency. If you want to control the weight of the car - the aerodynamics - you would progably need a one or two hertz (cycles per second) system. If you want to control the dynamics of the car (i.e., braking, acceleration and cornering) you need a system that is between six and 10 hz. If you want to control hub motion, you need a system of 20-30 hz. It is really a question of selecting the frequency for the response you need."

It is interesting to note that many notebook (under 4 lbs) micro-computers with Intel processors advertised in the Bay Area for prices under $2500 (with display and keyboard) use 386 CPUs that run at 33 to 40 hz. Obviously, the technology is readily at hand.

More from Wright, "Once you have a computer controlling anumber of aspects of the suspension system, it is very easy to change them, and even to allow the driver to change them. My understanding is that the Williams system allows the driver to change the roll stiffness distribution, as well as the front and rear ride height. Those are three of the most powerful car set-up parameters."

The Williams cars seem to have an added dimension, they are capable of rising up while on the straights, thus improving top speed. "My guess," Wright says, "is that the computer senses in some way that it is on the straight, or it changes the attitude of the car above a certain speed. That is really very easy. It is the very first thing you can do with any form of active control system."

"We have been working on it (active suspension) since 1981, when we ran an active road car," says Wright. "We have a number of people who are used to this technology; we have a good data base, and we have a lot of technology. It isn't easy in the first, early stages. You lose a lot of sleep, and your hair goes gray..."

It is obvious that a decade of develop at Lotus has made active technology seems ordinary. Meanwhile many F1 teams are working hard and scrambling to catch up with Williams and Lotus. Trouble is experience and development are virtually impossible to gain in short time.