Active Suspension 2021 - active spring mounts

[Tommy Cookers]

active suspension can be realised simply by using active spring mounts
(removing from conventional suspension the effects of relatively steady forces eg aerodynamic downforce )

active spring mounts can be realised simply by ......
mechanical 'irreversibility' MI (or nearly that) or by electrical 'irreversibility' EI (or nearly that)
'irreversibility' means non-backdriveability ie the variable spring mount height behaves like a fast-acting jack
MI or near-MI works with conventional electric actuation ie electric jacking

electrical irreversibility EI can be realised from an (unusual) motive device having very uneven flux distribution
the stepper or variable reluctance principle
this can be designed to reach and easily hold any of a number of discrete positions
positions of very high flux density at which a small supply of electrical energy will hold against external 'backdriving' force
or the position selection can even be moved by electrical energy but held by permanent magnetic flux only

it's simple and cheap
it's green (could generate its own power)
so it's road-relevant

what's not to like ?

remember
a servo-hydraulic active suspension system consumes energy continuously except at steady load ie when the car is stopped
a conventional ie 'voice coil' electric active system consumes energy continuously
this system as above consumes at steady load no energy at steady load and at varying loads consumes energy intermittently

[johnny vee]

Nice post man, very informative.

Admittedly I know very little about active suspension. I remember that the Williams system in 1993 was full on actuators and solenoids that also handled rebound and damping whereas the Lotus (Adrian Newey) had springs and dampers and that whole system was adjusted hydraulically to keep car level in braking and acceleration.
That is if I remember correctly.

2021 could be awesome!

[Jolle]

Nice post man, very informative.

Admittedly I know very little about active suspension. I remember that the Williams system in 1993 was full on actuators and solenoids that also handled rebound and damping whereas the Lotus (Adrian Newey) had springs and dampers and that whole system was adjusted hydraulically to keep car level in braking and acceleration.
That is if I remember correctly.

2021 could be awesome!

There is a big difference between active suspension F1 style, active suspension road style and ride hight stability.

Active suspension mounds are ride hight and that is the system that lotus used (not with Newey by the way, it's one of the teams he never worked for). They could rise or lower the car depending on windspeed.

Road cars have a combination of ride hight and stiffening up the suspension. The springs and dampness still do all the work. For racing this system would be relevant if you could stiffen up suspension independently and constantly.

The F1 Williams style active suspension was really cool, no springs, no dampers only a big ass pump and a piston on each wheel with a computer pulling or pushing a wheel up and down. This was also it's downside, there was no backup. If you would loose pressure or the computer would crash, all the wheels would be up in the air.

[johnny vee]

Thank you for that Jolle

[Maritimer]

Movable spring seats and magnetorheological dampers would achieve everything they would need no?

[AngusF1]

Movable spring seats and magnetorheological dampers would achieve everything they would need no?

Hmm, that depends on what they 'need'. If possible engineers would want total control of ride height, springing, damping and inerting on each corner at all times, to any values they wanted, arbitrarily quickly. The closest to that would presumably be a full active system.

[Maritimer]

Movable spring seats and magnetorheological dampers would achieve everything they would need no?

Hmm, that depends on what they 'need'. If possible engineers would want total control of ride height, springing, damping and inerting on each corner at all times, to any values they wanted, arbitrarily quickly. The closest to that would presumably be a full active system.

True, I mean more from Liberty's side. The dampers are pretty widely used already so tech is cheap(ish) and spring seats would give night adjustment. Whether or not they want systems that alter setting multiple times per second or just with longer movements like long turns or braking is another matter.

[Tim.Wright]

If I remember correct, the williams system had an air spring in parallel with the hydraulic actuator in the first iteration to carry the static weight force. This way the actuators didn't need to be powered to hold the car off the ground at standstill. I read that it was then changed to a belville washer stack because of problems with the non linear force characteristics inherent in air springs. A system like this needs to be extremely high bandwidth as it is responsible for low frequency load transfer/areo effects as well is high frequency road inputs.

A low frequency spring perch could be implemented with a soft parallel spring element to react the static weight loads, that way you don't need an irreversible mechanical system which typically introduce large amounts of friction and inertia (effective unsprung mass) to the suspension. which are the principal characteristics which limit the system bandwith, especially in an electro mechanical system.

[johnny comelately]

Overall i am suspicious that it will contribute to less overtaking by making the cars corner harder via better aero and moments/ vectors things ... it brings up the problem of convergence re car speed, one line, parity = delta

[AngusF1]

Absolutely, the suspension behaviour will become much closer to ideal across the whole range of conditions on all the cars, and the ability of a higher skilled driver to drive around undesirable behaviour will become less of a factor in lap time.

The move to active suspension runs completely counter to the stated goals of the new regs, which are to reduce costs and increase the driver's contribution to performance. In that context I don't understand it at all. It will require teams to develop new active systems and reduce the driver's contribution.

A ruleset in line with the goals of the reg's could have been to allow only one spring and damper per corner, with the only interlinking being one anti-roll bar at either end.


That said, from a technological development point of view full active systems would be very cool. Computer control systems and active valve tech have advanced at least two full generations since active was last tried in F1, and no-one else has optimised active in circuit racing since then. It would be an engineer's dream to have a whole new field to develop.

[johnny comelately]

Absolutely, the suspension behaviour will become much closer to ideal across the whole range of conditions on all the cars, and the ability of a higher skilled driver to drive around undesirable behaviour will become less of a factor in lap time.

The move to active suspension runs completely counter to the stated goals of the new regs, which are to reduce costs and increase the driver's contribution to performance. In that context I don't understand it at all. It will require teams to develop new active systems and reduce the driver's contribution.

A ruleset in line with the goals of the reg's could have been to allow only one spring and damper per corner, with the only interlinking being one anti-roll bar at either end.


That said, from a technological development point of view full active systems would be very cool. Computer control systems and active valve tech have advanced at least two full generations since active was last tried in F1, and no-one else has optimised active in circuit racing since then. It would be an engineer's dream to have a whole new field to develop.

and that = $

[DaveW]

and that = $

True, but the budget for the Lotus system installed in the ‘99T was rather less than 10% of AN’s annual salary for 3 systems designed, built and supported for the season.

Tim, I think you are not quite right. The Lotus system used “helper” springs, but the Williams system used air springs (like Citroen). US patent 4861066 contains Frank Dernie’s description of one version. Ultimately, that was the successful version (or, I like to think, was on the successful car). It had rising rate springs, but no active control over damping. There was a fundamental difference between the two systems: as downforce changed, the Lotus system consumed no (additional) power, but the Williams system consumed power to maintain ride height (if that makes sense).

If I may say so, Tommy’s post described the variable spring perch that has been in use for several years. Even the “stepper motor” approach has been used, I think successfully, but not quite in the same way he described. I think that Tommy’s scheme could be made to work if the ride height was “scheduled” using time, speed & track position. Using measured load directly would present too many problems. The think that Maritimer’s idea was a good one, but MR actuators would be rather too heavy for F1.

[Tommy Cookers]

easy things to say .......

in a spring based suspension isn't a feedback position signal (eg corner ride height) as good a corner load feedback signal ?

electromagnetic damping has more versatility than hydraulic and can recover all of its energy and more (ok its heavy etc)

an electromechanical equivalent of eg the system on the Lotus F1 could imo work up to c. 4 Hz
isn't that enough ? (given that car wheels have upstream pneumatic springs called tyres)

[Tim.Wright]

I found a post a made some time ago about the active ride systems that Adrian Newey has been involved in:

Some notes on the Leyton House system from the December 2012 issue of Motorsport magazine (Article on Adrian Newey)

• They started work in 1989.
  
  Was a "platform control system, which is purely aimed at trying to control ride height through speed and downforce variation instead of trying to deal with road inputs, and left conventional springs and dampers to deal with those"
  
  "The system we used on the Leyton House was very simple. You put an actuator under the spring platform, and compensated the tyre and suspension squash by extending the actuator.
  
  This type of system was used by McLaren in 1993.

Some notes from the March 2012 issue of Motorsport magazine (Article on the FW14B)

• The active system was originally proposed by AP in 1984: "The AP system managed ride control of a 'three-legged' platform via valves controlling the flow of high pressure hydraulic oil, the valves being controlled by linkage to the suspension and a parallel inertia-based system"
  
  First test car was made in winter 1985 based on an FW09. The active suspension was then put on the back burner for a season because the passive car was looking like being competitive in 86. An active test car based on the FW11 was tested pre-season in 87.
  
  Newey arrived in 1990 and continued the development of the three-legged system depite being of the opinion that the platform control concept from Leyton House was a better system.
  
  "[the Williams system] basically had 2 front springs and a single rear spring. And then the bump side of the single rear actuator was connected to the rebound side of the front actuator and that's what gave you the roll stiffness."
  
  In 91 they moved from gas springs to disc springs which resolved the high speed bouncing on the tyre sidewalls due the non linear stiffness of the gas springs (at high aero load they were too stiff).
  
  This system had a fixed front/rear roll stiffness distribution, so the balance was adjusted actively using the ride height.

in a spring based suspension isn't a feedback position signal (eg corner ride height) as good a corner load feedback signal ?

Suspension position doesn't contain information regarding damping and suspension jacking forces. Though these can be estimated pretty well if you know the vehicles braking, powertrain and suspension characteristics.

[DaveW]

electromagnetic damping has more versatility than hydraulic and can recover all of its energy and more (ok its heavy etc)

Do you have an example of an electromagnetic damper design complete with energy recovery?