The other active suspension cars - Leyton House

[PlatinumZealot]

Well... the reason why I asked because I deal with pneumatic valves/cylinders a lot in the industry I am in and I was just trying to relate the it to the active suspension. Sometimes we use double acting cylinders/valves but most of the time single acting works OK. The single acting ones only the "pushing" fluid circuit is controlled and for the pull they are spring returned. The double acting ones both the push and pull is controlled. The double acting one of course you have more control since you can control the deceleration to a finer degree and of course you can control the return force. In the case of active suspension I wondering if this is done too or they rely on the springs in the suspension for the return.

[PlatinumZealot]

I the drawing it is not clear if the cylinders are double acting or single, though I will take your word for it.

(Single acting cylinders can have the same number of ports as double acting)

[PlatinumZealot]

Are you sure it is double acting?
Might not need to be. The primary load usually acts in one direction. Hard to see the system needing to "pull up" a wheel.

Peter Wright has written a book entitled "Formula 1 Technology", illustrated by (the enviable) Tony Matthews. Appendix C is entitled "Active Suspension". With apologies to both, I have taken the liberty of copying a page taken from this Appendix.

Figure C1 illustrates a passive suspension (with all the "extras" omitted).

Figure C3 illustrates the Lotus "parallel" active suspension, with the damper of the passive system replaced by a true double acting hydraulic actuator (the ports are connected directly to an EHSV). If the parallel spring is designed to carry (roughly) the dead weight of the vehicle and the aero, then it is clear (hopefully) that the actuator must be double acting.

Figure C2 illustrates what Peter describes as "semi-active", & I have referred to earlier as a "series active system". Tony has drawn the actuator as "double acting", but Dernie's patent shows that the rear suspension actuators are "single acting", but the front suspension actuators are "double acting". In Dernie's patent, the spring of Figure C2 is the compressed nitrogen, and the damper is explicit (see figure 3). Arguably, the actuator of figure C2 is neither single nor double acting in Dernie's case, because it acts as an hydraulic rocker controlling the distance between sprung mass and the top of the suspension in Tony's schematic.

The differences between the two systems are interesting. On a flat road, and with a changing down force, the series system must consume energy to maintain ride height, but the parallel system does not. On the other hand, with road inputs, the series system will consume little energy, but the parallel system must consume energy to maintain control (if that makes sense).

Ok. What I see here is the "semi-active" implies ride-height control only. Figure C2. My interpretation is this system is the parallel one. Damping is working in parallel with ride-height control because they are two in different fluid circuits. Despite that the damper and actuator are visually connected in series.

The full active diagram, figure C3 implies active ride height and possibly active damping control. That the damping system is on board, and the same fluid that is used for damping is also the motive fluid used to control ride height.
I interpret this as series... quite opposite to you and I may be wrong.. but the fact that the motive fluid experiences the same forces as the damping fluid make me think it is series.

[DaveW]

Ok. What I see here is the "semi-active" implies ride-height control only. Figure C2. My interpretation is this system is the parallel one. Damping is working in parallel with ride-height control because they are two in different fluid circuits. Despite that the damper and actuator are visually connected in series.

The full active diagram, figure C3 implies active ride height and possibly active damping control. That the damping system is on board, and the same fluid that is used for damping is also the motive fluid used to control ride height.
I interpret this as series... quite opposite to you and I may be wrong.. but the fact that the motive fluid experiences the same forces as the damping fluid make me think it is series.

Apologies for my inexactitude, but I hope we now understand each other (with aid of the diagrams). F1 designers make a comfortable living by (mis)interpreting the FIA rules.

I quoted part of your last post, because it seems to me that you may have been too simplistic. There is a difference between pneumatic and hydraulic actuation...

The actuator shown in figure C3 appears to be working in parallel with the spring. That is not the case, however, because the actuator controls the spring position (completely, if the system is well designed), and the software controlling the actuators simulates both the spring and damper rates and the spring pre-load - and many other things. In fact the Lotus system emulated "modal" springs and dampers, heave pitch & roll "ride" heights, warp offset to control lateral balance, "pre-view" (kind of), and stroke limitation algorithms (think bump rubbers).

The actuator shown in figure C2 (logically) must emulate a negative spring (opposing the motion of the physical spring). That would undoubtedly be unstable, so it must also emulate a substantial damper - which is why the system cannot be perfect (riff_raff).... To be fair, the system was unlikely to be implemented in precisely that way, and the WGP system evolved with time (autogyro)... and I am as much in the dark as most, although some of the Citroen systems (see above) may provide a few clues.

[riff_raff]

Looking at the picture linked of the three systems, the only difference I can see between C2 and C3 is that C2 uses a dampener parallel to the spring. Both C2 and C3 appear to show a double acting actuator. You should also remember that the opposing force provided by a conventional suspension spring increases in compression and decreases in extension, and may not match the force/travel characteristics needed by the suspension. But the double acting hydraulic actuator will always be capable of regulating the suspension position regardless of opposing forces.

[DaveW]

I'm sorry, riff_raff, I really do not know how reply to your last post, other than to suggest you read the text of my last 2 posts. Unkind, I know, but there is not much else to say.

Other readers should be aware that the last sentence in riff_raff's post is not necessarily true. Power consumption of both systems depends upon actuator area, which also sets the actuator stall load, assuming constant supply pressure. Allowing the actuator to stall occasionally may be an efficient option overall. Also the loads reacted by the actuators of C2 & C3 are quite different, all other things being equal.

[PlatinumZealot]

Is it normal to have two dampers in series as what is implied in diagram C2?

At first I took it as ride height control only as Newey Said the Leyton house car's active suspension was only intended to make a stable aero-platform and not really for suspension management... ??

[Tim.Wright]

Is it normal to have two dampers in series as what is implied in diagram C2?

At first I took it as ride height control only as Newey Said the Leyton house car's active suspension was only intended to make a stable aero-platform and not really for suspension management... ??

As I understand its not 2 dampers in parallel. The top one is a double acting hydraulic actuator. The middle one is the suspension damper and the lower one is the effective damping in the tyre. Based on what I have read this is the system used in the Leyton House.

Because the passive and active elements are in series, they share more or less the same load. But their displacements sum. I.e. +5mm on the actuator and +5mm of the suspension gives a total ride height change of +10mm. So in this way the active system is working as an "offset" to a passive system.

The advantage of this is that the actuator only has to take care of primary ride (low frequency, high amplitude) and it leaves the passive system to take car of the high frequency low amplitude stuff where active systems tend to run into bandwidth problems (i.e. they can't react fast enough).

The low frequency stuff is generally easily predictable i.e. roll and pitch due to braking and cornering and vertical loading due to aerodynamics. So controlling these movements is relatively straight forward.

High frequency stuff is generally a lot more random like curb strikes, road asperities, shock loads from gear changes etc so they are difficult to control in a proactive way. Leaving these inputs to a passive system is quite an elegant solution.

[DaveW]

.... is quite an elegant solution.

Indeed would be, as described.

Life becomes more complicated if changes in weight and aero are to be tracked (particularly with DRS). Those might be tracked by including spring position in the feedback loops, but that could (would) cause instabilities when movement of the "preload adjuster" changed the vertical load, which would cause the potentiometer output to feed back spring position changes into the control system.

Not impossible to stabilize, but it does make the system more complicated.

[riff_raff]

I don't think you have fully considered what is described in the C2 schematic. It describes a downforce acting on a sprung mass, attached to a double-acting hydraulic (?) linear actuator, attached to a compression spring working in parallel with a double-acting hydraulic linear dampener, attached to an un-sprung mass, which is connected to a pneumatic rubber tire having its own peculiar spring rate and dampening characteristics.

You state that the advantage of the arrangement shown in the C2 schematic is that the hydraulic actuator only needs to operate at lower frequencies. I don't see how this situation would be the case since the actuator would end up constantly trying to adjust for the attached passive spring and dampener movements. If the active hydraulic actuator does not have adequate frequency response to keep up with the passive system motion, then I don't see how this approach would work effectively.

Developing a high frequency hydraulic suspension actuator was not that difficult a challenge at that time. So the approach shown in Fig. C3, with fewer variables, seems way more logical.

[Tommy Cookers]

bandwidth/frequency response etc ....
I(I assumed) some posts referring to electro-hydraulic valves meant electro-hydraulic servo valves
iirc historically there was terminology eg electrohydraulic valves, eh proportional valves, and eh servovalves
and some similar implied spread of cost and performance capability within the hydraulic actuation
I imagine the currently popular term Moog valve means a leading brand of eh servovalve

anyway, .....
a precedent system that by design use less than the most capable eh servo-hydraulic hardware is interesting
in part because today such a system could (and would ?) use electromechanical actuation not hydraulic actuation
technically that would have been possible even in those days

[DaveW]

I don't think you have fully considered what is described in the C2 schematic...

Making progress, riff_raff, but still a way to go...

a precedent system that by design use less than the most capable eh servo-hydraulic hardware is interesting in part because today such a system could (and would ?) use electromechanical actuation not hydraulic actuation
technically that would have been possible even in those days

We have had this discussion before, Tommy, (http://www.f1technical.net/forum/viewto ... 43#p524843). I think the real problems with your proposal are weight & packaging. Moog has published a technical bulletin (150) where, in Table 7, EM & EH actuation systems are compared.

[djos]

Looks like active suspension is coming to road cars as a 3rd party upgrade. Basically it's an active shock and still uses springs like the Leyton house system.

http://arstechnica.com/cars/2015/07/the ... in-action/

You do need a CanBus equipped car tho.

[Tim.Wright]

You state that the advantage of the arrangement shown in the C2 schematic is that the hydraulic actuator only needs to operate at lower frequencies. I don't see how this situation would be the case since the actuator would end up constantly trying to adjust for the attached passive spring and dampener movements. If the active hydraulic actuator does not have adequate frequency response to keep up with the passive system motion, then I don't see how this approach would work effectively.

This actuator will do more or less what the control tells it to do - and if you tell it to only move at (relatively) low speed then that's what it will do. In fact, you could block this actuator solid (0Hz) and the suspension will still work.

Low frequency body movements which come from braking, accelerating and cornering (which occur mainly in the range of 0-5Hz) can be calculated pretty reliably based on longitudinal/lateral acceleration. So there is also no need to include the damper displacements in the high level control loop of the active system. This avoids the stability problems which Dave mentioned before.

[riff_raff]

This actuator will do more or less what the control tells it to do - and if you tell it to only move at (relatively) low speed then that's what it will do. In fact, you could block this actuator solid (0Hz) and the suspension will still work. Low frequency body movements which come from braking, accelerating and cornering (which occur mainly in the range of 0-5Hz) can be calculated pretty reliably based on longitudinal/lateral acceleration. So there is also no need to include the damper displacements in the high level control loop of the active system. This avoids the stability problems which Dave mentioned before.

Please explain how the "low frequency" active hydraulic actuator would be able to keep pace with movements of the "high frequency" passive suspension it is attached to. The digital controller can send command signals to the valve actuator at whatever frequency it desires. But that does not mean the hydraulic actuator will be able to keep up with the movement demanded by the controller.

There was some discussion of the difference between proportional valves and servo valves. But this is mostly an academic distinction. Electro-hydraulic valves are differentiated by both the frequency and flow rate required. The frequency and flow rate required by a full active F1 suspension system can be handled by digital servo-valves that have been available since the late 60's from companies like Sturman.