F1 active suspension

[DaveW]

They could be made to work safely, but in modern terms the AS from that era wouldn't fly today on safety grounds.

I would expect a proper AS system now would have magnetic fluid instead that can controlled at each corner damper, instead of a central hydraulic system. It could be made safe, it just wasn't then.

If the cars natural position was raised, and the AS did the lowering, then a failure would mean the driver could brake and pull off, or finish the race if it was drive-able.

Some things that would be interesting with a rethink would be that we no longer have dampers in the rear, just heave and ARB. AS would mean going to a less tightly packaged rear to provide controllable dampers in the rear corners. This might negate some of the aero gained over recent bulimic purges over the years.

What I didn't like about AS is took away much of the sense of speed for the viewer and driver. The changes of direction looked like nothing, as if it required no thought, just a video game like crisp line.

Stated as facts. However, I have major issues with every single statement & assumption in the above quote, I'm afraid.

[Pingguest]

Conclusion
Considering all, there is no reason not to re-legalize active suspension. Let engineers show us what they can do!

There's one very fundamental reason, the same reason it was banned, saftey.

It's why they keep having to slow the cars down year after year with more and more restrictive regulations...

That's the reason why I think geomatric limits shouldn't be tightened continuously and engineering masterpieces shouldn't be banned, homologated or standardized. Instead, performance parameters (e.g. the amount of downforce) and consumables (e.g. number of tyres per race) should be limited.

[Giblet]

They could be made to work safely, but in modern terms the AS from that era wouldn't fly today on safety grounds.

I would expect a proper AS system now would have magnetic fluid instead that can controlled at each corner damper, instead of a central hydraulic system. It could be made safe, it just wasn't then.

If the cars natural position was raised, and the AS did the lowering, then a failure would mean the driver could brake and pull off, or finish the race if it was drive-able.

Some things that would be interesting with a rethink would be that we no longer have dampers in the rear, just heave and ARB. AS would mean going to a less tightly packaged rear to provide controllable dampers in the rear corners. This might negate some of the aero gained over recent bulimic purges over the years.

What I didn't like about AS is took away much of the sense of speed for the viewer and driver. The changes of direction looked like nothing, as if it required no thought, just a video game like crisp line.

Stated as facts. However, I have major issues with every single statement & assumption in the above quote, I'm afraid.

Fair enough, care to elaborate? The sled thing is true, the old system having a single point of failure is true, everything else I was speculating and trying to make conversation and encourage discussion.

[hardingfv32]

I was under the impression that the fully developed F1 AS systems did not have dampers or ARBs.

Brian

[GSpeedR]

We should not confuse an "adaptive" suspension system with an "active" system. An active suspension directly inputs energy into the system from an external supply (in most cases, a hydraulic pump). If desired, an active system can apply force in the direction of a velocity (rather than opposing velocity). An adaptive suspension is essentially a passive system that can change how energy is dissipated or stored, but it cannot input additional energy. MR fluid and electro-servo valved dampers are adaptive; they can continuously modify damper characteristics and they do have a power supply, but they only change dissipation rates. Hence, I would not label an MR damper systems as "Active".

Active systems can be placed in parallel with a passive system to provide a load path under a failure. Bump-stop can be (and are) used to keep the chassis off the track enough to avoid certain catastrophe. Considering the wheel rates of F1 vehicles, this probably wouldn't be too difficult to implement.

I could see the benefits of both types of systems, though the costs of adaptive systems should be far less. Full decoupling of suspension modes (active system) could allow for better control of mechanical grip while keeping the aero lords appeased.

[DaveW]

Fair enough, care to elaborate?

Let's try.

The Lotus system installed on the 99T (the first active car to win a race) had parallel springs & bump rubbers that were capable of supporting the vehicle, & the hydraulic actuators that normally over-powered these to control the vehicle acted as crude dampers when the system was depressurized. The system was designed specifically so that the suspension would function after a power failure. As I stated earlier, Senna finished a race in that state, completing 6 laps (& losing around 10 seconds per lap, I recall). Hardly a "sled".

I don't think that MR fluid dampers could be considered to be an "active suspension". They have independent hydraulics, true, but I'm not aware that they have ever been used to control right height. They do have a single point of failure, the controller. Also, if current is not pumped continuously to each corner, the actuators revert to a nominally undamped state. The dampers are serious heavy - hardly an F1 contender, I think. A lighter solution is available (in theory), using ER fluid, but that requires seriously high voltages (circa 700 V DC). Overall, then, I would question the idea of "rheological" fluids as "active", they have implementation problems, they don't intrinsically improve reliability, & the failure mode is not ideal....

I have yet to encounter a decoupled damper solution on the rear of an F1 vehicle. In fact, there are good arguments against the idea.

I think the major cornering speed change happened with skirted venturis. Skirts had been banned by the time first AS car ran (1983). I cannot think that drivers thought they where playing a video game....

[Giblet]

Thanks Dave.

I see what I thought was terribly incorrect about some of the systems and their operation.

Unfortunately though the sled analogy was not exactly made up, a car bottomed out when its AS failed. A body with no wheels or useless wheels is a sled to me. It is uncontrollable and high speed.

At Spa-Francorchamps for the Belgian Grand Prix Alessandro Zanardi was heading into the ultra-fast Eau Rouge section when a leak in his Lotus’s hydraulic system caused the bottom of the car to hit the track. The car went straight on into the barriers at enormous speed.

http://www.f1fanatic.co.uk/2007/05/17/b ... uspension/

So his hydraulic system failed, the car bottomed, and the wheels were useless.

You wil find I don't make things up, and I usually have a reason for saying them.

[xpensive]

With some mechanical engineering creativity, which still xist I've heard, there is possibly a way to build in some redundancy- or default-system to prevent the car from falling to the ground completely even if hydraulic pressure is suddenly lost?

[Giblet]

With some mechanical engineering creativity, which still xist I've heard, there is possibly a way to build in some redundancy- or default-system to prevent the car from falling to the ground completely even if hydraulic pressure is suddenly lost?

Yeah, which was the whole reason I posted what I did earlier... I was just thinking out loud, albeit feebly, that AS in modern F1 it might need a rethink?

Would hydraulic be best still? Pneumatic?

I think if the the suspension remain traditional, as to me car setup is very important in judging teams and drivers. If the roll was controlled actively, it could keep the car level and low for the aero and stance, not dissimilar to the MP4-12C.

[GSpeedR]

It is certainly possible to have "protection" solenoids located on the return chamber that default to the closed position; they would require a power signal to open (after which they no longer affect the system). If safety checks are tripped or electrical power is lost then the solenoid shuts, and your ride spring is the fluid compressibility, assuming no leakage past the piston/ram. If that is too much weight/complexity then put a bump-stop on the shaft. The Zanardi incident is an example of poor implementation of AS with no (?) mechanical safety features present.

Pneumatics versus hydraulics is a question of control strategy. For systems that need to be "tuned hot" to provide rapid response for platform control and disturbance rejection, you probably want a hydraulic system with servo-valves. For systems that simply require very low frequency platform control and can absorb disturbances with no/little rate change and little damping (albeit with more mid-high freq motion), pneumatics may be used. F1 teams would desire the former I would guess.

Pneumatics absolutely would have to be in parallel with a traditional suspension to provide damping and stiffness. Air is an awful working fluid and considering the scale of these actuators, there would be tremendous amounts of hysteresis and friction. Not to mention there is no way they can package accumulators large enough to provide adequate capacitance. Pneumatics are excellent for slow-speed ride height changes (on your SUV), and with enough capacitance they are analogous to high-preload low-rate springs.

[WilO]

Great summary of implementing a proper failure mode to a technology that would add much to F1 (and road cars (in my opinion), albeit for different reasons). Nice contribution GSR, as usual.

[DaveW]

You wil find I don't make things up, and I usually have a reason for saying them.

Touché... It is interesting to read somebody else's interpretation of events. e.g.

These early attempts at active suspension were more ‘reactive’ – using hydraulics to alter the car’s attitude in response to bumps in the road or particular inputs from the driver rather than actually preparing the car in advance for each specific change in the track (That was always the case up to the end of 1987). The first two iterations of the system (the second driven by Nigel Mansell occasionally in 1983) did not even use springs. (actually, the 1983 system was the first iteration actually designed for F1, & spring support was planned, but the first solution failed on test.)

The first system was abandoned when Renault refused to allow a hydraulic pump to be attached to it's engine (rather than lack of data processing capability).

Williams, however, were developing a system of their own ("Frank Derney's" system) which was more limited in the scale of its ambitions than Lotus’s (As far as I am aware, it was based on the Citreon hydropneumatic system). It consumed less power (about 5bhp) (actually the Lotus system was not that dissimilar, but was continuous) and Piquet gave an active Williams its first win at Monza – after Senna had gone off (True - but he took to the sand over-taking a back marker whilst leading the race fairly comfortably).

Actually, by the time we understood the vehicle it was pretty competitive (in Senna's hands) for a time. Later, it lost relative performance & it became obvious that the system was (no longer) working its tyres hard enough. Senna rejected the obvious solution & softer tyres were not available (for some reason).

I ceased to be closely involved with the system after the middle of 1987, although my group Cranfield continued to design & supply controllers. Hence I can't say in detail why the AS accident occurred. Perhaps the pressure of weight saving caused them to take risks, or perhaps, the set-up was wrong for the massive vertical acceleration of Eau Rouge. After all, BAR did manage write off two passively suspended cars in quick succession some 10 years later, presumably caused by inappropriate set-up.

[DaveW]

Pneumatics versus hydraulics is a question of control strategy. For systems that need to be "tuned hot" to provide rapid response for platform control and disturbance rejection, you probably want a hydraulic system with servo-valves. For systems that simply require very low frequency platform control and can absorb disturbances with no/little rate change and little damping (albeit with more mid-high freq motion), pneumatics may be used. F1 teams would desire the former I would guess.

A good summary, GSpeedR. Actually, I believe that pure Pneumatics would simply not be powerful enough or fast enough to cope with the rapid changes in downforce that must be reacted in an F1 vehicle. A hydro-pneumatic system, using air springs with ride height controlled hydraulically & suitable damping devices is attractive, but a compromise still has to be made, choosing a leveling time constant that achieves close (enough) ride height control without fighting the springs & dampers (too much).

The Citreon system (which doesn't have large D/F changes to cope with) handles the compromise by reacting to a ride height error only after a significant delay.

[DaveW]

Not really important, but I recently unearthed a photograph I took during the second test of the Lotus 92 Active Car at Paul Ricard.

I had forgotten that the failure of the initial attempt to provide back-up suspension was following quickly by a revision of the rear installation to accommodate coil-over helpers. Hopefully, you can see those in place on this shot of the vehicle.

For interest, the photograh shows (from left) Steve Hallam (race engineer), Steve Green (development engineer), Nigel Mansell (being fitted), fixer extraordinare Bob Dance (holding up the rear wing), Peter Warr, Peter Wright. On the right, Mr Hicks (truckie) is explaining all to Ferrari truckies (pit lane intelligence).

Not shown is the unsightly telemetry arial we used at the time to gather data (citizens band radio was the best we could afford - it worked unfailingly down straights, but had a tendency to break up in corners due to reflection from the stands).

[WilO]

Great photo for a number of reasons Dave, thank you for sharing.