Porsche 919 Decoupled Suspension

This is a mirror of a very good article written by Andrea Quintarelli Porsche 919 Front Suspension Part1

A more generic article about the car in Pdf format is mirrored here Race Car Engineering Porsche 919 2014

After a long absence from sportscar racings top level, and a very convincing come back in 2014 with the new 919 Hybrid LMP1, Porsche succeeded in winning Le Mans at their second attempt. The 919 went on to dominate the following two years.

Beside many other interesting technical aspects and technical secrets, the 919 has shown from its first debut a very intriguing concept for both front and rear suspensions. Not too many pictures are available to analyze the details of Porsche suspensions, but it is clear that they look pretty different than any competitor traditional arrangement, at least concerning dampers and springs function and activation.

Porsche front suspension architecture already caught attention in 2014, as the car debuted in the LMP1 class. The car changed a bit in 2015 when a slightly different layout was chosen for springs, dampers and antiroll bar. This same scheme was adopted in 2016 and 2017, although some minor updates were introduced from year to year to further refine system’s performance (including, most probably, an interconnection between the front and rear suspension).

Porsche 919 front suspension – Source: www.movetenms.com

Porsche 919 front suspension. Source: www.racecar-engineering.com

Porsche 919 front suspension. Source: www.roadandtrack.com

But what do 919’s suspensions have that other cars don’t? And what do we know about them?

The main source shedding some light (in German) about the principle behind what (probably) Porsche uses is a patent that anybody can find on the internet and that describes a suspension system able to decouple roll and heave control, both in terms of stiffness and damping.

I described the above mentioned patent and my interpretation of Porsche 919 2014 rear suspension in my Blog (DR Racing Blog), building up in CAD some models of what shown by the available sources and simulating their kinematics in a pure heave and roll motion.

Here, we will try to look at front suspension scheme used from 2015 on instead, using some 3D CAD models to reproduce what is shown in the available pictures and describe in detail how we think Porsche suspension works.

With the patent the only official source coming directly from Porsche, we will again start taking a look at its content, mainly trying to understand how could operate the system that it describes. Interesting enough, the document was published in 2009, so some years before Porsche come back in LMP1 and at a point in time where Porsche RS Spyder LMP2 cars (latest sportscar work program, before 2014) was still on the track, although heavily penalized by latest rule changes.

The idea of decoupling roll and heave control is not new for race cars: there are many pictures available proving that many Formula 1 teams tried this pattern, with different means to reach their scope. Already in 2009, Williams was probably using no corner springs on their car’s rear suspension, where a third spring was responsible for controlling vertical movements and an antiroll bar was resisting any roll motion, with still three shock absorbers to dampen vertical and roll motions.

Among LMP1 cars, Peugeot used for years a rear suspension layout in part similar to the one we will describe in a while, although they probably never took the corner springs completely out.

Actually, some of the readers who are familiar with pushrod/pullrod layouts with T antiroll bars could object that, even with such a “traditional” suspension layout (widely used in many minor Formulae) a stiffness control decoupling could be theoretically achieved, removing the corner springs (similarly to what probably Williams did), using a third element unit (placed on top of the T-bar) to counteract heave movements and an antiroll bar to control roll motion. The corner dampers could be tuned to dampen appropriately in roll, while a third damper could be used to contribute to further dampen any vertical movement of the body.

F1 rear suspension with torsion springs, corner dampers and third spring/damper unit – Source: www.carthrottle.com

In any case, any other layout is still a compromise, in comparison to the one used by Porsche. To properly control roll and heave motions, both in terms of stiffness and damping, Porsche 919 employs only two dampers and two springs, allowing for lower weight, an easier package and a very compact layout, although some other small components are used to link the different parts together.

Moreover, as we will see, Porsche’s layout allows each unit (damper and/or spring) to take care of only one function (roll or heave), while other traditional layouts would have at least the corner dampers acting both in heave and roll.

The reasons why a race engineer dreams about completely decoupling heave and roll control are pretty easy to understand: on one side, that would allow to practically adjust only the suspension mode you desire, with no compromises of any sort; besides this, an interesting (secondary) effect would be having a chance to dampen “correctly” both roll and vertical motion.

One of the compromises that need to be accepted using a traditional layout is that the same dampers are responsible for both heave and roll damping. That means, the corner shock absorbers have to deal with vertical movements but also with roll motion, with the two modes normally requiring different tuning. In the author’s experience, if the corner dampers are “correctly” tuned for heave, the roll mode is normally overdamped. We could discuss if this is necessarily a bad thing and if a proper roll damping was the prime aim of Porsche, but this is out of the scope of this article.

Briefly, from a purely mechanical perspective, an overdamped system is slower to react, but some drivers could like the feeling given by a very strongly damped roll motion because they have the impression of being better supported by the car.

What is clear is that, with a system that could allow a (nearly) ideal separation of roll and heave control, the engineers would be free to decide how much damping to apply to each mode (for example, still keeping roll damping coefficients pretty high, if preferable) without affecting any other suspension mode.

Porsche seems to have solved this dilemma in a very elegant way, keeping the layout pretty neat and simple, still having a decoupled control of roll and heave motions and leaving the door open for an easier interconnection of front and rear suspensions (FRIC) acting only on the heave unit, as they probably did.

Although it doesn’t represent a realistic race car suspension design (probably for confidentiality reasons), the above-mentioned patent is very useful to show how decoupling roll and heave could be done and which main components are needed. Its main picture is an overview of the concept itself:

Porsche Patent

Here below, a picture of a 3D scheme of (our interpretation of) what is shown by the patent:

A number and a colour are assigned to each component. Part n.1 and n.2 (Red and purple lines) are the lower and upper control arms respectively. Part n.3 (Black line) is vehicle’s chassis, while parts n.4 are drop links connecting the Lower Control Arm protrusion to the elastic elements. In pure heave, they move a Watt Linkage (part n.5, light blue), free to rotate around a vertical axis, defined by a yellow vertical bar (part n.6); this bar is free to sway around a horizontal axis.

This part is actually not shown in Porsche’s patent and it is not strictly necessary to let the system work as described, but its addition doesn’t take any generality out of the concept and it is one of the possible interpretations to describe how the roll elements could be activated.

Between one side of the watt linkage and one of the lower control arms protrusion seats a Spring/Damper unit, which only works in Heave (part “A”, initial length 225mm). On the other hand, the yellow vertical bar activates, when swinging about its horizontal axis pivot (only in roll) a “roll damper” (part “B”, initial length 225 mm) and, through a drop link (part n.7, orange) an antiroll device, directly mounted on the chassis (part n.8, brown), here shown in the shape of a bending blade.

In a pure heave motion (both wheels moving up with respect to the chassis), the vertical bar (part n.6) doesn’t sway, but the watt linkage rotates about its vertical axis and let the “Heave Spring/Damper” unit (part “A”) to compress or extend (as shown in the following picture), where its length reduces to about 160 mm.

In a pure roll motion, the watt linkage doesn’t rotate and the “Heave Spring/Damper” unit is not active. The complete upper part of the system moves rigidly and no elastic/damping element is involved in any motion in this area.

The Vertical Bar (part n.6) sways about its pivot and activates both the roll damper unit (Part “B”) and the antiroll bar (or, maybe better, antiroll spring, part n.8). In the picture below, the first has now a length of about 239.8 mm (the simulated roll motion is pretty big, about 50 mm wheel motion per side or about 3.6 degrees, which is unrealistically big for a similar race car, but helps to better underline the principle and one of the few areas of concern of this concept, namely the roll damper motion ratio; more about this later).

The antiroll spring could also be incorporated in the pivot axis of part n.6 (the yellow vertical bar), in the form of a torsion spring. We stuck to what shown in the patent, though, for sake of clarity.

This simple 3D model shows how the patent’s concept would grant to control roll and heave separately, both in terms of stiffness and damping.

Now follows Part 2. Porsche 919 Front Suspension part2

In the first part of this analysis of Porsche 919 front suspension, we analyzed how a suspension system like the one described in a Porsche patent and aiming to decouple roll and heave control could actually work.

As I mentioned, one area of concerns could be to achieve a roll damper motion for each degree of roll big enough to ensure an effective damping and avoid the undesired influence of damper internal friction (although latest technology dampers, using much lower internal gas pressure, probably reduced already internal friction).

Let’s now take a look at what the actual 919 front suspension (could) look like. As for the patent scheme, we rebuilt the suspension layout in CAD in order to show how its kinematic could work. Now, please note the model obviously doesn’t have the right dimensions, but should only help to explain Porsche’s front suspension working principle, above all in term of springs and dampers activation. Also, it doesn’t represent a working design, above all in terms of components stiffness and strength: it should only serve as a visualization aid to explain how the mechanism works (a picture is worth a thousand words, after all). Finally, as I said, what we are going to show here is only our taking about how Porsche’s scheme could look like; Porsche is most probably not keen to reveal any of their secret to us, so we cannot be totally sure that what we sketched is absolutely right.

Again, a number and a colour are assigned to each component. Parts n.1 (dark grey) are lower and upper wishbones, Part n.2 (pale brown) is the tie rod, Parts n.3 (orange) the two pushrods and Parts. n.4 (grey) are the two rockers. The pushrods move the rockers, which rotates around a pivot axis represented by the blue bracket attached to the chassis. When moving, the rockers activate a mechanism that works on a very similar principle to the one described by the patent. More specifically, one rocker is connected to a rod (Part n.5, pink) and the other one to the heave spring-damper unit (Part n.8, red) and to a second link (Part n.7, purple). Let’s now take a closer look at the central components of this layout.

The main element, here, is the Central Rocker (Part n.6, yellow), which pivots around an axis fixed to the chassis (protrusion in its lower part) activating, most probably, a coaxial torsion spring that works as an anti-roll device. Its upper extremity is directly connected to the roll damper (Part n.9, light blue) that, as we will see, only moves in roll.

As in the patent’s scheme, the central rocker is not activated directly by the rockers and the lateral links. The latter are actually connected to a central lever, functionally very similar to the Watt linkage we have seen in the patent’s scheme. It can be better seen in the following picture where, for clarity, we removed the yellow central rocker to make the inner components visible:

As we can see, both the pink (part n.5) and purple (part n.7) links activate a central lever (part n.10, dark green) which is free to pivot around an axis fixed to the central rocker. This lever is also connected, on its upper part, to a short rod (part n.11, dark purple) which then activates the other side of the heave spring-damper unit (part 5, in red).

Please note that the design of the short rod (part n.11) and of the central lever (part n.10) are here just a sketch to show how each component is activated and represents in no way a real working design. We know they would need to build in a different way, above all in their common interface, to ensure proper stiffness and strength. A similar scope could be achieved, for example, using a ball joint between part n.5 and part n.10 and an axial bearing between part n.10 and part n.11. But such a detailed study is out of the scope of this article. Each and every 3D model shown in our pictures only wants to be an aid to explain how the kinematics works.

The reader can probably see how similar this scheme is to the one shown in the patent, although it was only a simplified layout to describe the working principles of Porsche’s concept.

Of course, the control arms protrusions of the patent are here gone and have been replaced by a double wishbone setup with pushrods and rockers. Still, the key elements of this layout are the central rocker, taking the place of the sway vertical element of the patent, the central lever, that replaces the watt linkage, the heave spring-damper unit and the roll spring and damper.

We can now take a look at how this layout behaves when the suspension moves. We will analyze two situations: a pure heave motion, with a jounce (body moving vertically, downward in our case) of 25 mm and a pure roll motion of an angle of 1 degree. For clarity, the central rocker has been made partially transparent in following pictures, to let the reader see what happens to the central lever in both situations.

Also, the heave spring has been removed, since its deflection would anyway be equal to the one of the heave damper, being the two coaxial. Finally, the length of both the roll damper and the heave damper will be shown, to analyze how much they deflect.

Let’s first start with the design condition:

As we may see, in this condition the heave spring-damper unit is 260 mm long, while the roll damper is about 300 mm long.

If we now consider a heave motion, we would a see a displacement in the heave unit while the roll damper maintains approximately the same length.

As shown in our picture, the roll damper nearly doesn’t move, staying at a length of about 300 mm. The heave damper displaces, going to a final length of about 224 mm. The system we built up in CAD is not perfect. The roll damper has here a deflection of 0.023 mm, which confirms it is possible to achieve a nearly perfect decoupling.

As we can see in the picture, in a pure heave movement, the central rocker doesn’t practically move. The central lever, instead, rotates about its pivot on the central rocker and, through the short rod, compresses the heave damper which is, on the other side, also pushed by the rocker.

Since the central rocker doesn’t move, this is also the case for the roll damper, on top of it.

Key to making all of this happen is the motion ratio of the central lever and of the corner rocker. Referring to the next picture, this means that the ratio between the distance of point A to point B and the distance of point B to point C must be the (close to) the same of the ratio between the distance of point D to point F and the distance of point E to point F.

Considering a roll motion, on the other hand, we would see the components moving as shown in the following picture:

Here, the heave damper nearly doesn’t move (we talk here of a 0.17 mm displacement over a 1-degree roll suspension movement, which is unrealistically big for such a car) while the roll damper is now activated and displaces pretty significantly (about 21 mm). How much the roll damper moves directly depends on how far its attachment point on the central rocker seats from the pivot.

As the reader can see, now the central lever doesn’t move at all. As we said, our CAD model doesn’t produce an ideal separation but the error we have is in the region of parts manufacturing tolerances and can be neglected. Also, we can assume that Porsche put much more effort than we did in optimizing their kinematics to extract the maximum out of it.

This simple study shows what should have been the working principle of Porsche front suspension (rear suspension was using a similar principle, but achieved it with a different layout, as explained in my Blog) and why it is so interesting. As we mentioned, Porsche engineers developed an extremely elegant and neat system to completely decouple roll and heave motion, ensuring not only tuning advantages but also a very compact package and the use of lower numbers of components.